The effects of Chinese lending on the economic policymaking in developing countries : does an access to Chinese loans lead to less neoliberal policies?

(1)

Master Thesis Political Science

Specialisation Track Political Economy

The Effects of Chinese Lending on the Economic Policymaking

in Developing Countries:

Does an Access To Chinese Loans Lead to Less Neoliberal

Policies?

By Tadas Jakeliūnas

Student ID 12274356

Supervisor: Dr. Julian Gruin Second Reader: Dr. Farid Boussaid Words Excl. Bibliography: 17675 June 2019

(2)

1. Introduction

Before the 21st century, developing countries and emerging market economies were dependent on development finance from international financial institutions (IFIs) such as the IMF and the World Bank, as these institutions were among the few actors that were able to finance infrastructure projects crucial to development. However, the financing does not come without a cost, as loans, grants and credits from these IFIs usually include stringent conditions to pass policy reforms. The conditionality often includes economic policy reforms which are sometimes called neoliberal (Rodrik, 2002; Öniş and Şenses, 2005; Stiglitz, 2008; Babb and Kentikelenis, 2018). Neoliberalism is not a monolithic concept and some argue that it is used differently in different contexts and by different scholars and therefore it has lost any substantive meaning (e.g. Boas and Gans-Morse, 2009).

However, if we are to use a simple, yet substantive definition of neoliberal policies that is used by political economists, as "any measure intended to lessen the role of states and enhance the role of markets" (Babb and Kentikelenis, 2018), the links between IFI conditionality and neoliberalism become clear. Since the late 1970s, the IMF and the World Bank have for a long time been advocating, and are still advocating to a large degree today, policy instruments of the Washington Consensus, such as deregulation of markets, privatisation of public enterprises and economic austerity measures (Williamson, 1990). Most of the policies of the Washington Consensus, especially those that were mentioned, transfer the power from the state to the market forces. Therefore, IFIs are one of the main transnational actors that help institute neoliberal reforms across the globe and in developing countries in particular.

In the 21st century, however, the global environment of development financing began to change. Possibly the biggest development has been the rise of China. Since the 2000's, China has transformed itself from almost a non-actor in development financing to one of the main financial partners of developing countries all around the world. In the year 2000, China loaned only 130 million USD to countries in Africa and provided no loans to Latin American nations, while in 2016 Chinese development banks provided over 30 billion USD in loans to African states (Atkins et al, 2017) and over 21 billion USD to Latin American countries (Gallagher and Myers, 2019). The World Bank, which in the past was the largest development financer in the world, committed around 46 billion USD in loans, credits, grants and guarantees in 2016 (World Bank, 2016a). Thus, the 2010's will be the first decade since the Marshall Plan in which a nation state, instead of an IFI, will become the largest development financer in the world.

(4)

4

The rise of China in development financing has grasped the attention of international relations (IR) and international political economy (IPE) scholars. The aspect of Chinese lending that raises the most questions for scholars is the fact that, unlike the loans from IFIs, Chinese loans usually come "with few or no strings attached" (Condon, 2012: 6, also see Bosshard, 2007; Wu and Wei, 2014). Among the "few strings attached" are foreign policy conditions, such as a political commitment to the One-China policy, conditions to back credit lines with natural resources and conditions to hire Chinese contractors to implement the development projects that the financing is to be used for (Mattlin and Nojonen, 2015). These conditions have made Chinese lending an interesting research topic for scholars who are interested in the motivations and strategies China has with its lending (e.g. Dreher and Fuchs, 2011; Chen, Dollar and Tang, 2016; de Morais, 2011)

Thus, when scholars write about "few or no strings attached", they usually mean that China, unlike IFIs, does not include economic policy reforms as part of its conditionality. Even though financial agreements between Chinese development banks and its borrowers are not publicly available, the absence of economic policy reform conditions in Chinese lending seems to be a consensus among scholars. While the existing Chinese conditions on foreign policy and natural resource extraction have led scholars to ask questions about Chinese lending motivations and strategies, the effects of the absence of Chinese economic policy reform conditions on the political economy of developing countries has not yet been thoroughly studied.

The lack of economic policy conditions in Chinese lending is particularly important when the dominance of IFIs in development finance in the decades prior and the stringent neoliberal conditionality of IFIs is taken into account. Before the rise of China, developing countries that wished external financing for their development and infrastructure projects, had to accept IFI conditionality, which lessened their policy space and lead to neoliberal policymaking. Now that there is an option for developing countries to instead get financed by Chinese development banks, this might change.

First, in the most obvious way, developing countries might just switch from IFI financing with policy conditionality to Chinese financing without policy conditionality, and in this way maintain full economic policy autonomy while being able to finance their development projects. Of course, Chinese resources are not limitless, therefore this might not always be an option and sometimes developing countries would still have to enter financial agreements with IFIs even when they would not desire the conditionality that comes with the financial agreements. Furthermore, some governments might fully trust the knowledge of IFI experts and view the financial agreements as a way to pass reforms that they themselves deem necessary, albeit not necessarily popular among the public.

(5)

5

However, the rise of China and its non-conditional lending might give developing countries more policy space indirectly as well. Because loans without economic policy conditions are probably more attractive than those with conditions, IFIs, in an attempt to compete with Chinese development banks, might be forced to make their own loans more attractive. One way to achieve this is to also include less economic policy conditions. Therefore, IFIs might make their conditionality less stringent when entering financial agreements with developing countries that also borrow from China. Thus, in theory, developing countries are no longer required to sign financing agreements with IFIs that include stringent, neoliberal conditionality every time they want to start an ambitious infrastructure project, as now they can look in China's direction and obtain financing while maintaining autonomous economic policymaking. Furthermore, IFIs are forced to make their own conditionality less stringent in an attempt to compete with China. All of this should give developing countries more policy space. But has this development actually lead to less neoliberal economic policymaking in developing countries? This is the question that is going to be answered in this thesis.

(6)

6

2. Literature Review

Since structural adjustment programmes became a part of the IMF and the World Bank lending practice, the effects of development finance on the political economy of policymaking in the borrowing countries has become an important research field in IPE. The aspect of IFI lending that probably received the most interest from scholars was the often stringent conditionality of financing agreements, as it required the borrowing countries to enact economic policy reforms that otherwise might not have been enacted. Furthermore, while some economists view the reforms included in the conditionality of the IFIs as non-ideological, sound economic policies (e.g. Burnside and Dollar, 2000), others have called it market fundamentalist or neoliberal (Stiglitz, 2008). This means that the major actors in development financing pushed for policy reforms that reflected one ideological view of development while alternative paths existed.

Therefore, much research has been done on how the conditionality of the IFIs affected economic policymaking of borrowing countries. The scholarly debate on this topic originates in the field of economics, when Burnside and Dollar (2000) found that while there is no correlation between development aid and growth in general, development aid significantly increases growth in "good" (good for growth) policy environments and concluded that development aid should include more conditionality to increase growth in developing countries. Easterly, Levine and Roodman (2003) then used a different dataset to test this hypothesis again and found that there was no relationship between development aid and growth in "good" policy environments, which lead them to the conclusion that further research should rather test whether conditional developmental aid leads to "good" policies themselves.

Since then, a number of scholars tried to tackle this question. Several scholars tried to find a relationship between IFI lending and economic indexes which measure the scope of economic liberalism in countries. Boockmann and Dreher (2003) found that the effect of development finance on economic freedom depends on the type of financing and the IFI that provides the financing: World Bank projects had a positive effect on economic freedom of the recipient country, World Bank credits had a negative influence and IMF programmes had no effect. Heckelman and Knack (2009) found a mixed relationship between development aid and economic freedom: it had positive effects on some categories and negative effects on others. Finally, Smets and Knack (2014) find a significant link between World Bank lending and two clusters of the World Bank's Country Policy and Institutional Assessment (CPIA) rating, which reflects the policy prescriptions of the Washington Consensus, however the relationship diminishes with each cumulative conditional loan.

(7)

7

More research has been done on the effects of the IMF and the World Bank lending on specific economic policies in the borrowing countries. Financial and capital account liberalisation was found to be positively affected by lending from the IMF by many different scholars (Abiad and Mody, 2005; Chwieroth, 2007). Lending from the IMF also was found to lead to deregulation of FDI inflows (Vadlamannati and Cooray, 2014) and trade liberalisation (Weymouth and Macpherson, 2012). However, no relationship was found between the World Bank lending and trade liberalisation (Brooks and Kurtz, 2007). Bjørnskov and Potrafke (2011) looked at how the World Bank projects and the IMF programmes affected privatisation in postcommunist countries and found a positive relationship between IMF programmes and privatisation. Furthermore, Henisz et al. (2005) found a positive effect of both the World Bank and the IMF lending on market-oriented reforms in telecommunications and electricity industries. Kilby (2005) found that World Bank lending promotes deregulation of markets. Heckelman and Knack (2008) found a reverse relationship - according to their research, development aid slowed market-liberalising reform in developing countries, however their dataset did not include data on most of the IMF and World Bank lending.

Thus, the findings by different researchers show mixed results, however they do not seem contradictory and show that the effect of development finance on economic policymaking depends on the context: the donor institution, the type of financing, and the policy area affected. Overall, the links between IFI lending and neoliberal economic policies seem to exist in a lot of contexts, as the research that showed different results did not include World Bank and IMF lending in its data. World Bank lending and project lending, instead of credits, seem to have a bigger influence on indexes that display the scope of economic liberalisation as a whole. However, certain policy areas, such as trade liberalisation and privatisation seem to be more affected by the IMF lending.

While the effects of IFI lending on economic policymaking in developing countries has been thoroughly researched, the rise of development financing from national actors, especially China, raises new questions. However, since this is quite a new phenomenon, research is still lacking. So far, more research has been done on the motives and strategies China has with increasing its role in development financing (e.g. Dreher and Fuchs, 2011; Chen, Dollar and Tang, 2016; de Morais, 2011). The effects of the increased role of China in development financing on the political economy of policymaking in developing countries is a less researched area. In this research field, so far there are more questions than answers.

Thus, while the questions about the impact of IFI conditionality on policies of borrowing countries have mostly been answered, the research has not caught up with the rise of China in development finance. The world of development finance that was lead by IFI lending and their conditionality seems to have resulted in more neoliberal policies in developing countries. Now that

(8)

8

China has become a major actor in development finance, the effects of non-conditional lending should be researched as well. Therefore, in this thesis the question "does an access to Chinese loans lead to less neoliberal policies?" is raised.

(9)

9

3. Theory

This paper rests on five major theoretical claims that will help raise hypotheses and possible answers to the research question that was raised. First, the Washington Consensus policy instruments, which is a major part of the conditionality of IFIs, is, for the most part, neoliberal. Second, the Washington Consensus can be viewed as a policy paradigm, which is reliant on legitimisation from economics experts and coercive power, which belongs to IFIs. Third, we are entering a post-Washington Consensus era of development finance which will not be dominated by a single policy paradigm or a single transnational actor. Fourth, the post-Washington Consensus era came about not because of a change in IFI ideology; the IFIs are still subscribed to neoliberal worldview. Finally, empirical facts surrounding Chinese development banks show that they have a qualitatively different approach to development finance ant have the financial means to provide developing countries more policy space, and the emergence of these actors is the factor that truly challenges IFIs and creates a post-Washington Consensus era. In the next five sections of this chapter, each of these theoretical claims will be explained in more detail. In the final section, hypotheses will be raised according to the theory.

3.1. The Washington Consensus and Neoliberalism

The Washington Consensus became prominent during the Latin American debt crisis in the late 1970s and the 1980s. When Latin American nations were unable to pay off their debts, they turned to the IMF for new loans and help with debt restructuring. The IMF saw economic imbalances in Latin American countries and offered to help only if they agreed to implement economic policy reforms that would fix these imbalances. These conditions most often included policies that would decrease fiscal deficits, cut down or reprioritise public expenditure, broaden the tax base, create market-determined interest rates and competitive exchange rates, liberalise trade and FDI flows, privatise public enterprises, deregulate labour and goods markets and ensure property rights, and these conditions were so wide-spread that Williamson (1990) called all of these policy instruments a part of a consensus that exists among all Washington based institutions. Over time, Washington Consensus conditionality-based lending became a part not only of the IMF's crisis management practice but also a part of development financing, as the World Bank, regional banks and the IMF itself began including Washington Consensus conditions in development lending (Babb, 2013).

Washington Consensus policy instruments are often called neoliberal or market fundamentalist (Rodrik, 2002; Öniş and Şenses, 2005; Stiglitz, 2008; Babb and Kentikelenis, 2018). What makes the Washington Consensus neoliberal? First, it is important to define neoliberalism in a clear way. Since 'neoliberalism' is a term that is being used a lot and in many different contexts, it

(10)

10

has been described as a term without substantial meaning and with conflicting definitions (e.g. Boas and Gans-Morse, 2009). That is partly correct, as many different characteristics of neoliberalism are put forward by scholars and a lot of them are not universally accepted. For example, some argue that neoliberalism is different from other forms of liberalism because it paradoxically requires substantial state intervention to enforce free markets (Hickel, 2017). However, others apply the term to liberalisation of the economy without explicit state intervention. Another characteristic that is used to differentiate neoliberalism from other liberalisms is that proponents of it pursue marketisation of services that never before functioned as markets, for example by making public education and healthcare enterprises operate under the profit motive. Yet, the term is also applied to outright privatisation of public enterprises as well. Thus, while these are popular claims about what neoliberalism entails, they are not universally accepted.

A simple, yet substantial and concrete definition is offered by Babb and Kentikelenis (2018), they define 'neoliberal policy' as a "any measure intended to lessen the role of states and enhance the role of markets in at least one national economy". This definition of neoliberalism might seem not exclusive to neoliberalism, as different forms of liberalism, such as laissez-faire liberalism may be defined in a similar fashion. This might be true to an extent, especially because the same economic policy trend that started in the 1980s is sometimes called by names different than neoliberalism, including laissez-faire and market fundamentalism (Stiglitz, 2008). Yet, neoliberalism is the term that is used most often and the definition that is given by Babb and Kentikelenis (2018) is both concrete enough to not become a meaningless term and broad enough to encompass many different forms that neoliberalism has taken since the 1980s. Therefore, most of the Washington Consensus policy instruments can be defined as neoliberal, especially such as privatisation, deregulation of markets and cutting down public expenditure. Furthermore, over time Washington IFIs started to include neoliberal reform conditions that went beyond the original Washington Consensus, for example privatisation of public water utilities, privatisation of public pensions systems, replacement of progressive taxation with value-added taxes and capital account liberalisation (Babb, 2013).

3.2. Policy Paradigms and the Washington Consensus

The role of IFIs in establishing the Washington Consensus was already touched upon in the last section, however it is important to embed the relation between IFIs, the Washington Consensus and neoliberalism in a more concrete theoretical framework. This is especially important because some scholars treat the Washington Consensus more as a consensus between leading economists, and therefore more as an intellectual product instead of one embedded in international relations actors

(11)

11

and institutions (e.g. Rodrik, 2006; Stiglitz, 2008). This categorisation of the Washington Consensus as a scientific paradigm may lead to believe that the spread of the Washington Consensus and neoliberalism happened only because it was legitimised by economists, with no political legitimisation and coercion by international political actors.

Instead, Babb (2013) believes that the Washington Consensus should be treated as a transnational policy paradigm. A policy paradigm is a "framework of ideas and standards that specifies not only the goals of policy and the kind of instruments that can be used to attain them, but also the very nature of the problems they are meant to be addressing" (Hall, 1993, pp. 279). Policy paradigms, like scientific paradigms are legitimised by scholars, however, unlike scientific paradigms, they are also embedded in practices of bureaucracy and other organisations and legitimised by political actors (Ibid.). This means that policy paradigms can stay paradigmatic not only because of almost unilateral legitimisation from experts but also because the paradigm gains coercive power because political or bureaucratic actors can coerce other actors to comply with the paradigm through power (Babb, 2013).

The Washington Consensus paradigm was put into place mainly through two mechanisms. First, through normative means, as neoclassical economics became the mainstream in the 1970s and legitimised more market-friendly economic policies (Babb, 2013). Second, governments, especially those of developing countries, were coerced by IFIs, as accepting their conditions often was the only way to borrow in case of crisis and to finance development projects (Ibid.). The second, coercion mechanism, is the more important in the context of this paper. As will be shown in the following sections, the Washington Consensus is no longer the consensus among economics scholars, however coercive pressures from the IFIs are still in place. Therefore, the IFIs become the main maintainers of neoliberal policymaking in developing countries. This in turn means that the biggest challenge to neoliberal policymaking in developing countries would have to come from outside the IFIs, possibly in the form of new lenders and donors in the world of development finance.

3.3. Post-Washington Consensus

As mentioned in the previous section, IFIs had a specific role in the maintenance of the Washington Consensus policy paradigm, a role of coercion. The two biggest IFIs, the IMF and the World Bank, as well as some of the most important regional development banks, are known for their conditions to pass policy reforms that are part of financial agreements. These institutions also had committed the most resources to finance crisis management and development projects. Thus, for a long time, the world of development finance was led by international actors that maintained the neoliberal policy paradigm.

(12)

12

However, since the mid 2000s and, even more so, since the 2010s, a wide range of scholars began suggesting that the neoliberal Washington Consensus era of development finance is either coming to its end or is already over and the Post-Washington Consensus era is beginning (e.g. Rodrik, 2006; Stiglitz, 2008; Babb, 2013; Grabel, 2017). Among these scholars, there are several different viewpoints on how and through what mechanisms the Post-Washington Consensus manifests itself. Some scholars look at the Post-Washington Consensus as a change in the consensus among mainstream economists on what are good economic policies, for example Stiglitz (2008) argues that the new consensus is that one-size-fits-all policies are no longer considered good and that the role of the state should not be minimal, as was under the Washington Consensus. However, as was established in the previous chapter, policy paradigms, such as the Washington Consensus, are not solely maintained through expert legitimisation, but also by bureaucratic tradition and coercion. Thus, a change in what top economists view as good economic policies is not enough if IFIs still maintain the Washington Consensus through coercion with conditionality-based lending. As will be explained in the next section, this is true not only for top economists that do not work for IFIs but also for economists from within IFIs themselves. Even a more critical view on neoliberalism from economists in IFI research departments does not guarantee a change in IFI practice, as a cleavage exists between IFI economists and the political dynamics of IFI lending.

The second viewpoint is that the Washington Consensus is over because the most important actors that maintained it, the Washington IFIs, have reformed themselves and no longer maintain it. Proponents of this viewpoint, for example Rodrik (2006) andBroome (2015), cite IFI rhetoric that is critical of their own past actions, less promoting of the Washington Consensus structural adjustment programmes and the increased attention of IFIs in the rhetorical promotion of institutional reforms, pro-poor spending and other reforms that were missing from the neoliberal Washington Consensus policy package. However, as will be shown in the next section, the changed rhetoric and the introduction of some non-neoliberal policies in their conditionality did not amount to much change.

The third viewpoint is that the era of the Washington Consensus and neoliberalism in the world of development finance is over, yet no new policy paradigm has come to replace it. For example, Grabel (2017, pp. 17) argues that the world of developmental finance, that has for a long time been dominated by neoliberalism, is entering a state of "productive incoherence". What distinguishes this era from the neoliberal era is the "absence of a consensus around any particular unified theoretical ideal toward which institutions of financial governance are to hew" (Grabel, 2017, pp. 16). Under neoliberalism, there was a general accordance between major state actors, IFIs and influential scholars on what development finance should look like. In the present, this coherence is lost. This "productive incoherence", according to Grabel (2017, pp. 17) might lead to more policy

(13)

13

space, more experimentation in developing countries and might provide them the opportunity to "chart alternative economic trajectories". Grabel (2017) sees some change in IFIs, however admits that it is limited, and sees more potential to change the landscape in new actors, for example new IFIs from the Global South and rapidly rising national development banks from countries like China and Brazil.

A similar viewpoint is held by Babb (2013), who highlights that a new policy paradigm would replace the Washington Consensus only if it was legitimised by experts, embraced by a majority of policymakers around the world and enforced by transnational authorities, which is not yet the case. For this reason, the Washington Consensus will not be replaced by any policy paradigm, and a heterogeneous international regime will emerge (Babb, 2013). The Washington Consensus is no longer legitimised by experts and new international actors are beginning to compete with the IFIs for influence. Thus, the Post-Washington Consensus might manifest itself not as another consensus between top scholars and main international actors in development finance but rather as an era of competition in the field of economics and between the main international actors that have the coercive power to influence policies in developing countries. The latter competition is the target of this paper. In the next two sections it will be explained why Washington IFIs - the IMF and the World Bank - are still likely to coerce their borrowers to enact neoliberal policies, why Chinese development banks are the main competitors of the said IFIs, and why borrowing from Chinese development banks might lead to more policy space and, in turn, less neoliberal policymaking.

3.4. Continuation of Neoliberalism in IFIs

One of the viewpoints discussed in the last section sees the potential for the Washington Consensus to disintegrate through reforms in IFIs that were instrumental in its maintenance in the past (Rodrik, 2006; Broome, 2015). In this view, there are signs that the IMF and the World Bank are critical of their own past actions and that these institutions have rethought their conditionality moving forward. But has this actually happened?

On a rhetoric level, the change surely did happen. Rodrik's (2006) embrace of the World Bank came after the latter released a report critically reviewing its own policies in the 1990s, which was meant as a sign that the World Bank is separating from the Washington Consensus. The IMF, similarly, has come out with press releases warning about dangerous consequences of high income inequality and insufficient social protection policies (IMF, 2014; IMF, 2015). IMF economists Ostry, Loungani and Furceri (2016, pp. 38) even outright criticised the neoliberal model of development in IMF's own magazine "Finance & Development", saying that "instead of delivering growth, some neoliberal policies have increased inequality, in turn jeopardizing durable expansion". However, a

(14)

14

change in rhetoric is not equal to a change in policy. First off, rhetoric does not have to be sincere and reflected in practice, as it can be done for public relations purposes. Second, even if there is true a shift in the ideology of IFI economists and they change their opinions on certain neoliberal policies, this might not necessarily lead to a change in the practical behaviour of IFIs. Changing bureaucratic tradition is difficult in any organisation, and even more so in organisations where the educated opinions of researchers have to compete with the influence of many different political, state actors. Therefore, if neoliberal policymaking is to no longer be reinforced by IFIs, a change in their conditionality is a necessary condition.

Some parts of their conditionality have indeed changed since the early 2000s. First, the two Washington IFIs no longer push for policy reforms involving capital account liberalisation (Grabel, 2017, pp. 193; Kentikelenis, Stubbs and King, 2016). Furthermore, some non-neoliberal conditions, such as "pro-poor spending" have become a frequent part of their conditionality, although these conditions are often not implemented and their non-implementation does not lead to a discontinuation of financial support, unlike in the case of non-implementation of Washington Consensus reforms (Kentikelenis, Stubbs and King, 2016). Labour market liberalisation conditions and conditions associated with social security cuts have also sharply declined since the mid 2000s, although in cases where they have been included, they have been very stringent and extensive (Kentikelenis, Stubbs and King, 2016). Thus, perhaps the most extreme example of neoliberal policy is no longer a part of IFI conditionality and IFIs have embraced a few measures that increase the role of the state in comparison to the market.

On the other hand, many policy instruments associated with the Washington Consensus continue to be a part of the IFI conditionality up until today. In the case of the IMF, there was a slight decline in trade liberalisation, exchange rate policy and foreign direct investment policy reform conditions since the 2000s, however most countries dealing with IFIs have already liberalised their external sectors by the year 2000 (Kentikelenis, Stubbs and King, 2016). Furthermore, even though there was a slight decline in the inclusion of these types of conditions, they are still a part of the majority of financial agreements. The same goes for the privatisation of state-owned enterprises. Conditions to privatise them are still included in most programmes, even though the number of them has declined since their peak in the 1990s (Kentikelenis, Stubbs and King, 2016). As was the case with the liberalisation of the external sector, privatisation might have declined simply because after decades of IFI influenced privatisation in developing countries there are less publicly-owned enterprises left to privatise (Kentikelenis, Stubbs and King, 2016). There are less studies on the extent of the World Bank conditionality, however nearly 30% of the World Bank's portfolio since 2005 is made up of conditional lending and 19% of the World Bank's conditions are related to

(15)

15

privatisation (Babb and Kentikelenis, 2018). Furthermore, the access to the World Bank's lending is still allocated based on the CPIA rating (Babb and Kentikelenis, 2018), which, according to the World Bank's own researchers, "broadly reflects the so-called "Washington Consensus" neoliberal policy prescriptions" (Smets and Knack, 2014).

This means that developing countries still face policy space restrictions when dealing with IFIs. Furthermore, since IFI conditionality still largely reflects the Washington Consensus, IFI lending should still lead to neoliberal policymaking in developing countries. Thus, having an access to lending provided by a competing development finance actor that does not provide lending based on policy conditionality should provide developing countries more policy space, and, in turn, lead to less neoliberal policymaking. The next section will explain why China might be an actor that has the biggest potential to both compete with the Washington IFIs financially and why Chinese lending might give developing countries more policy space.

3.5. Chinese Development Banks and Policy Space

In this section, it will be explained why Chinese development banks in particular are of incredible importance in the changing world of development finance. Chinese development banks are not the only new institutions that are rising in significance in development financing. The Asian Infrastructure Investment Bank (AIIB) and the New Development Bank (NDB) formed by BRICS countries are both newly created development banks with substantial resources, furthermore the Brazilian Development Bank (BNDES) has substantially increased in assets over the last decade (Grabel, 2017). However, the NDB only operates in the BRICS countries and the BNDES mainly operates in Brazil, therefore they would not be able to influence policymaking in developing countries at large. While the AIIB has substantial capital and is said to become a rival IFI to the IMF and the World Bank, until 2017 the bank has only disbursed a little over 4 billion USD loans, and therefore is not yet a competitor to Washington IFIs.

Chinese banks, however, have both the resources to change the world of development finance and the qualitatively different approach to development finance than Washington IFIs. Two Chinese banks are of significant importance - the Chinese Development Bank (CDB) and the Export-Import Bank of China (Exim Bank of China). While the latter is not technically a development bank, it finances almost as much development projects as the CDB. Together, Chinese banks in 2016 provided over 30 billion USD in loans to African countries (Atkins et al, 2017) and over 21 billion USD in Latin America (Gallagher and Myers, 2019). These figures are already larger than that of the World Bank, the largest development financer in the world before the rise of Chinese development banks. Thus, Chinese development banks have the resources to lend significant amounts of money to

(16)

16

developing countries. Having these resources is crucial to change the landscape of development finance, as small amounts of loans would not make developing countries less reliant on lending from Washington IFIs, and, therefore, would not change the power dynamics.

However, having enough resources is not the only condition for a new actor to change the landscape of development finance. If Chinese development banks operated in the same way as traditional IFIs and applied the same conditionality in their finance agreements, the rise of these development banks would not have any effect on policymaking in developing countries, as they would only be able to choose their financier based on the size of the loan, the interest rate, etc. Therefore, having a completely different approach in terms of the conditionality of financial agreements is the second important thing when it comes to which new actors will be able to affect the way development finance is conducted and how much it affects economic policies of the borrowing countries.

In this regard, Chinese development banks also are fundamentally different from Washington IFIs. Some scholars argue that Chinese loans often come "with few of no strings attached" (Condon, 2012: 6, also see Bosshard, 2007, Wu and Wei, 2014). Others argue that Chinese conditionality includes foreign policy changes, for example a political commitment to the One-China policy, certain project-related demands, such as hiring Chinese contractors for the realisation of the project, as well as backing credit lines with natural resources (Mattlin and Nojonen, 2015). There are no reports, however, for economic policy conditionality. Thus, Chinese development banks take a completely different approach to traditional IFIs: IFIs often still include neoliberal economic policy reforms as part of their conditions for the financial agreement, while Chinese development banks do not require any economic reforms whatsoever. First, this practice should give more policy space to developing countries when they choose to enter financial agreements with Chinese development banks instead of IFIs. Furthermore, countries that have shown the ability to attract Chinese loans should have more bargaining power when dealing with IFIs. IFIs should try to maintain their relevance as development financers and compete with China. To make their loans more competitive, IFIs might make their conditionality less stringent and thus give more policy space to countries which also borrow from China.

3.6. Hypotheses

The theoretical claims laid out in the previous theory sections can be summed up as follows. First, the Washington Consensus is a mostly neoliberal policy paradigm. Second, the policy paradigm of the Washington Consensus is in a large part held up by the coercive power of the two Washington-based IFIs - the IMF and the World Bank. Third, we are entering an era of

(17)

Post-17

Washington consensus, where no single actor or policy paradigm should dominate development finance. Fourth, even though the Washington Consensus and neoliberalism in general is no longer the only economic policy ideology in development finance, Washington Consensus policies are still a big part of traditional IFI conditionality and they have coercive power to maintain neoliberalism. Finally, Chinese development banks offer lending without policy conditions and they have enough resources to become serious competitors to the Washington IFIs.

These claims lead to the proposition of a few different hypotheses. First, because Chinese development banks both offer lending that is not policy-conditional and provide a large amount of loans, developing countries that enter financial agreements with China should be less reliant on IFI lending. In turn, this should provide developing countries that deal with China more policy space. Because IFI lending has been shown to lead to neoliberal policymaking through Washington Consensus conditionality, this should also mean that developing countries that have access to Chinese lending should enact less neoliberal economic policies. It is not possible to empirically display a direct link between Chinese lending and policymaking itself. However, because developing countries should enact more neoliberal policies, this should lead to a change in macroeconomic indicators associated with the level of liberalisation of the economy. The Heritage Index of Economic Freedom is an indicator that reflects the liberalisation of the economy well, this will be explained in more depth in the chapter "Methods and Data". Thus, developing countries that borrow from China should score less on this index than those that do not, therefore:

H1. In developing countries that receive more Chinese lending, there will be a larger negative change in the score on the Heritage Index of Economic Freedom during the selected period than in those countries that receive less Chinese lending.

Because there is a lack of studies that researched the links between Chinese lending and economic policymaking, the confirmation of H1 is necessary to further argue for specific mechanisms that might cause this to happen. Without first establishing that developing countries differ in their macroeconomic indicators linked with liberalisation depending on the amount of Chinese lending they receive, it would be impossible to argue that any mechanism related with Chinese lending would cause developing countries to enact less neoliberal policies. After all, economic policymaking should be reflected in macroeconomic indicators.

On the other hand, a correlation between Chinese lending and liberalisation indicators would not demonstrate that Chinese lending causes less neoliberal policymaking. To test whether there is causation between the variables, a few more hypotheses about the causal mechanism will be put forward. The theory suggests two ways in which an access to Chinese lending might lead to less neoliberal policymaking in developing countries, one in which an access to Chinese lending should

(18)

18

directly influence policymaking in borrowing countries by replacing IFI lending, and one in which Chinese lending indirectly should indirectly change policymaking by improving the bargaining position of developing countries and making IFI conditions less stringent.

Figure 1.

First, because receiving loans from Chinese development banks would make developing countries less reliant on IFI lending, they might borrow less from the IMF and the World Bank and decrease the coercive power of IFIs. This should especially be true for policy reform-based lending, such as the World Bank's Development Policy Financing (DPF) programmes, which are more focused on policy reforms than specific development projects, and IMF's Extended Credit Facility (ECF), which focus most on structural reforms in non-crisis situations in exchange for concessional loans, unlike some of the other IMF programmes. This would mean that these countries would accept less conditions from the IMF and the World Bank, the conditionality of which are often neoliberal. The acceptance of less neoliberal conditions would in turn mean less neoliberal policymaking in that country. This causal mechanism will further be called the 'direct path' of the influence of Chinese lending on policymaking in the borrowing developing countries. Thus, two more hypotheses can be put forward that are necessary to test the direct path:

H2a. In a typical case, a developing country will participate less in IFI policy reform-based programmes during periods of significant lending from China.

H2b. In a typical case, development projects financed by China will be accompanied by less neoliberal policies than those financed by IFIs.

The second way that developing countries that borrow from Chinese development banks might gain more policy space and enact less neoliberal economic policies is due to the fact that IFIs might want to compete with China to retain their status as the most important development financers in the world. Furthermore, if IFIs want to maintain the Washington Consensus and believe that neoliberal policy instruments are the most beneficial for developing countries, they might still make

Access to Chinese lending

Less borrowing from IFI policy reform-based programmes Less reliance on IFI lending Less neoliberal policies enacted Lesser scores on Index of Economic Freedom Less neoliberal conditions by IFIs H1 H2a H2b H3a H3b

(19)

19

their conditionality less stringent. This is because IFIs might be forced to choose one of two scenarios. Either they lose a big part of their borrowers to China, or they include less conditions, make their lending more attractive and keep influencing policymaking in those countries, albeit in a lesser way. Thus, because developing countries that borrow from Chinese development banks are less reliant on IFI lending, IFIs might include less conditions in their financial agreements with these countries, therefore:

H3a. In a typical case, IFIs will include less neoliberal conditions for a developing country during periods of significant lending from China.

H3b. In a typical case, development projects financed by IFIs will be accompanied by less neoliberal policies during periods of significant lending from China.

(20)

20

4. Methods and Data

The research question of this thesis has not been investigated in previous studies, therefore in this thesis, both the correlation between Chinese lending and economic liberalisation and the mechanisms leading to different economic policymaking are studied. To investigate the causal chain of the process, a correlation between Chinese lending and macroeconomic indicators that reflect the extent of neoliberal policies in a country must first be established. If there is no correlation between the two, it is hard to imagine that the policymaking in countries with more Chinese lending is significantly different than in countries without it. Thus, a mixed method research design was used in the analysis. Quantitative methods are required to test H1 and a robust linear regression analysis was used for this. To test H2a-H3b, the process tracing method was used, with both qualitative and quantitative data invoked in the argumentation. In the next sections of this chapter, both methods and their application in this analysis will be explained in more detail, along with sample and case selection issues, time periods and data selection.

4.1. Robust Linear Regression Analysis

A robust linear regression analysis was used to test H1. The method was chosen because it is useful in testing the correlation strength between two qualitative variables. A sample of all Latin American and African countries with data available was used. The number of cases in model 1 is 61, and in model 2 is 58, as a few cases had data missing on one or a few control variables. The sample was chosen for a few reasons. First, all Latin American and African countries are classified as developing countries by the IMF. Second, data on the amount borrowed from China in most of the countries of both regions (Atkins et al, 2017; Gallagher and Myers, 2019) exists, while the same data on developing countries in other regions is not available. The data selection also dictated the selected time period of analysis. The years 2005 to 2017 were selected, as 2005 is the first year that data on how much Latin American countries borrowed from China is available and 2017 is the last year that the same data is available for African countries. A robust regression instead of a standard least squares regression was chosen because the number of cases included in the analysis is medium, therefore it is susceptible to outlier cases, which are not accounted for in the standard least squares linear regression. As the results of the analysis will show, there were indeed a couple of outlier cases, the effects of which on the analysis were counteracted by the robust regression method.

The independent variable in the analysis was the amount borrowed from China between 2005 and 2017 divided by GDP, as using this variable is more appropriate for this research question than using the absolute amount borrowed from China. This is because the same amount of money borrowed will affect larger economies less than smaller economies. For a smaller or less

(21)

21

economically developed country, a loan of 1 billion USD might satisfy all the needs for infrastructure and development projects, while the same amount might only cover a fraction of costs for a larger or more economically developed country. Thus, a larger country might still have to borrow from IFIs to finance all of their development projects and accept IFI conditionality. In a similar fashion, a 1 billion USD Chinese loan to a large economy might not be enough for IFIs to start competing with China and make their conditions less stringent.

There are many different macroeconomic indicators that could show the extent of neoliberal policies, with each indicator reflecting how much of a role the state and the market have in a specific policy area. It would be difficult to determine which of those should be included in the analysis and which ones should be left out. Furthermore, the results would definitely be mixed, as some areas might correlate with Chinese lending and others would not, which would mean that the hypothesis would not be properly tested. For this reason, the Heritage Index of Economic Freedom was used as part of the dependent variable. The use of an index which is calculated from many different macroeconomic variables allows to holistically test the hypothesis. The Heritage Index of Economic Freedom fairly well reflects the extent of neoliberal policies in a country, as its formula includes tax burden and government spending, business and labour market regulations, monetary, financial and fiscal regulations. Similar indexes were used in similar studies before (Boockmann and Dreher, 2003; Heckelman and Knack, 2009). The actual dependent variable will be equal to the Heritage Index of Economic Freedom in 2017 subtracted by Heritage Index of Economic Freedom in 2005. This will show how the overall economic policies of each country and the macroeconomic indicators that reflect them changed between 2005 and 2017 and to what extent they became more or less neoliberal in this time period.

Of course, economies of developing countries might become less liberalised due to other factors as well. Therefore, a few control variables were included in the regression to control for alternative explanations. One of the alternative explanations could be that China lends disproportionately more to authoritarian countries, which might be less supportive of economic liberalisation and neoliberal policies. Thus, a decrease in the Index of Economic Freedom in these countries might be explained simply because the countries are authoritarian and not because they borrowed from China. Therefore, the first control variable included is political regime, for which data from Polity IV was used. Another factor that is often associated with economic liberalisation is the level of economic development itself, as countries with larger GDP per capita are said to be more likely to liberalise their economies. Thus, the second control variable was GDP per capita (PPP) in 2005. The data used for this variable is from the World Bank. Third, the starting point of the level of economic liberalisation might matter, as, for example, the same amount of rise in the index of two

(22)

22

countries would be more substantive for a country that is less liberalised, and vice versa. Thus, the third control variable is Heritage Index of Economic Freedom in 2005. Finally, it is important to take into account the extent of the relationship between each country and IFIs. On the one hand, countries that have a deeper relationship with IFIs might be more likely to enact neoliberal policies. On the other hand, countries that borrowed more from IFIs in the past might have already passed extensive neoliberal reforms in the past and further liberalisation might be less likely. To account for this, the fourth control variable was the amount borrowed from IFIs (World Bank and IMF) before 2005 divided by GDP. The absolute amount is divided by GDP just like in the case of Chinese loans, as it is important to take into account the size of the economy of the borrowing country. Data on this was collected from the IMF and World Bank databases.

As was already mentioned, the results of the regression analysis only show a correlation and not causation. Furthermore, it is possible that the dependent variable would cause a change in the independent variable, and not vice versa. To account for this, a lag analysis could have been used, which would show which variable affects the other. However, time series data is required for such analysis and panel data was necessary to accomplish other objectives that were raised for the quantitative part of the analysis. Even without using lag analysis, two arguments can be put forward for the selected independent variable causing the dependent variable. First, logic dictates that it is more likely that borrowing from China would lead to less neoliberal policymaking by changing the relationship developing countries have with IFIs than it is that countries enacting less neoliberal policies tend to more actively borrowing from China. Second, the process tracing part of the analysis should show that this direction of the causation is indeed the case.

4.2. Process Tracing

While the robust linear regression analysis would show the correlation between Chinese lending and macroeconomic indicators associated with economic liberalisation, the correlation between the two variables would not automatically mean that there is a causal negative relationship between Chinese lending and neoliberal policymaking. Therefore, process tracing was used to show whether there is a causality between these variables and what the causal mechanisms are. H2a-H3b were tested using the process tracing method, while analysing a case of one selected developing country. In mixed-method research designs, it is important to analyse a typical case that is close to the regression line (Lieberman, 2005; Beach, 2018). Furthermore, it is crucial to select a case that has a high enough value of the independent variable, as the causal mechanism would not be present in a case where the cause itself is missing (Beach, 2018). As will be shown in the analysis chapter, the case that fitted these requirements the best was Cameroon, therefore this was the selected case.

(23)

23

With this part of the analysis, both qualitative and quantitative data was used to check whether the causal mechanism indicated in the theory part of the thesis and the hypotheses H2a-H3b are correct. The selected period for this part of the analysis was 2000-2017 instead of 2005-2017 as was in the case of the regression analysis, as it is necessary to include earlier data for the comparison of lending numbers and conditions. To test H2a, data on the IMF and the World Bank reform-based lending was compared against the data on Chinese lending. This was done to find out whether IFI policy reform-based lending decreased as Chinese lending increased in the selected case. To put these numbers into context, the overall change in IFI lending to the selected country was also overviewed, as were the global IFI lending numbers. The hypothesis will be considered confirmed if the selected country borrowed less from IFI policy reform-based lending programmes as it began borrowing more from China.

To test H2b, a counterfactual analysis was conducted. A variety of qualitative data was used. First, annual reports of China Development Bank and the Export-Import Bank of China as well as secondary literature were used to find out which development projects were financed in the selected country case. Finding out which policy areas the projects belong to is important when trying to figure out whether the policies that accompanied these projects would have differed if they were financed by IFIs. To find out how what conditions IFIs would have included in similar financial agreements and what economic policies would have followed, IFI financial agreements with other developing countries were used as counterfactuals. These documents show which conditions IFIs have included in similar cases with other countries when they were more reliant on IFI lending. It can be expected that similar conditions would have been laid out in the selected case if the country decided to borrow from IFIs. All these documents were compared with the actual policies and laws that accompanied China-financed development projects in the selected country. H2b will be considered confirmed if the actual policies relating with the policy area of the development projects which were financed by China are less neoliberal than the conditions identified in the counterfactuals.

For H3a, the selected country's financial agreements with the IMF and the World Bank were compared before the country started to borrow from China and afterwards. All the conditions were evaluated based on how much they required to liberalise and deregulate the economy. Then, the two periods were compared with each other. H3a will be considered confirmed if the selected country received less neoliberal conditions from the IMF and the World Bank after it began borrowing from China. To test H3b, the actual policies associated with the IFI financed projects before and after the country began borrowing from China were analysed. For this, laws passed by the country's parliament and the programme implementation evaluation documents by the IMF and the World

(24)

24

Bank were evaluated based on how much they liberalised and deregulated the economy. H3b will be considered confirmed if less neoliberal policies were enacted in the period after the country began to borrow from China than in the period where it did not receive Chinese lending.

The confirmation of H2a and H2b would confirm the presence of the first causal mechanism leading from an access to Chinese lending to less neoliberal policymaking, as it would show that Chinese lending made the country less dependent on IFI lending, which in turn led to less neoliberal conditions accepted and enacted. The confirmation of H3a and H3b would present evidence of the second causal mechanism, in which the access to Chinese lending affects leads to a change in IFI conditions. A confirmation of at least one of the causal mechanisms would show the effects the emergence of Chinese lending has on economic policymaking in developing countries.

(25)

25

5. Analysis

In this chapter of the thesis, the results from the analysis will be presented. First, the results from the robust linear regression analysis will be presented and interpreted. The results of the analysis will then be used to either confirm or disconfirm H1. Then, the results from the process tracing analysis will be presented. The direct mechanism of Chinese lending affecting policymaking will be presented first and H2a-H2b will be either confirmed or disconfirmed by the analysis. Then, the second, indirect mechanism of Chinese lending affecting IFI behaviour will be presented in the third section of the chapter and H3a-H3b will be either confirmed or disconfirmed.

5.1. Regression Results

Table 1.

As was indicated in the last chapter, the dependent variable of the analysis is the difference between 2005 and 2017 Heritage Index of Economic Freedom. The independent variable is Chinese loans received divided by GDP of each country. Model 1 of the analysis was a bivariate regression between these two variables. To account for alternative explanations for the liberalisation of the economy, in model 2, four control variables were introduced: the average Polity IV score of

(26)

26

countries, which indicates their positioning on the democracy/autocracy scale, the Economic Freedom index score in 2005, which indicates how liberalised their economies were at the start of the analysed period, GDP per capita in 2005, which indicates their economic development level at the start of the period, and IFI loans received before 2005 divided by GDP, which indicates the depth of the relationship between them and IFIs before starting to borrow from China.

Table 1 shows the results of the analysis. In Model 1, the relationship between the relative Chinese lending received and the change in the Economic Freedom index is statistically significant at the p<0.01 level. Therefore, there is a correlation between Chinese lending and economic liberalisation when unaccounted for other variables. After introducing the control variables in Model 2, it turns out that two of the control variables had no relationship with the dependent variable. First, Polity IV score of a country was unrelated with changes in the Index of Economic Freedom, which means that the economic liberalisation processes were similar in both more authoritarian and more democratic countries. Second, there was no relationship between GDP per capita in 2005 and the dependent variable, which indicates that the starting level of economic development was unrelated with further economic liberalisation in this sample.

However, the two other control variables correlated with the dependent variable. There was a negative relationship between the level of initial economic liberalisation and the change in economic liberalisation between 2005-2017. As was mentioned in the methods chapter, this relationship seems logical as there is less potential to further liberalise the economy in countries which are already strongly liberalised. A negative relationship was also present between initial IFI loans received divided by GDP and change in economic liberalisation. A relationship between the variables was expected, however it was unclear whether it would be positive or negative. It turns out that countries that had deeper relationships with IFIs in the past were the ones in which further liberalisation happened slower since the rise of Chinese development banks. The same logic that applied to initial liberalisation, applies here, as countries that borrowed more from IFIs in the past are more likely to have already passed significant neoliberal reforms and further liberalisation should be less likely. The relationship between both of these control variables and the dependent variable was significant at the p<0.01 level.

Even though the starting economic liberalisation position of countries and the initial relationship between countries and IFIS were confirmed as possible alternative explanations for further liberalisation, the introduction of these and other control variables into the model did not make the relationship between Chinese lending and economic liberalisation weaker, as it stayed statistically significant at the p<0.01 level. H1 stated that "in developing countries that receive more

(27)

27

Chinese lending, there will be a larger negative change in the score on the Heritage Index of Economic Freedom during the selected period than in those countries that receive less Chinese lending", and this was true even after accounting for alternative explanations. Therefore, H1 is

considered confirmed by the analysis.

Figure 2.

Figure 2 shows a graphical representation of Model 1. First, the right side of the graph shows that there were two outliers that received significantly more loans from Chinese development banks than other countries when the size of their economy was taken into account. These countries were the Republic of the Congo and Djibouti. The latter received this significant amount of lending mostly for the construction of the Addis Ababa-Djibouti Railway that was carried out by the governments of Djibouti and Ethiopia. Ethiopia is landlocked country almost all of the country's trade passes through Djibouti. Ethiopia is also an important trading partner of China. Therefore, this significant amount of lending can be explained by China's motivations to help improve infrastructure that is important for trade in one it China's trading partners. Republic of the Congo was one of the countries that borrowed the most from China for road and air infrastructure, as well as renewable

(28)

28

energy projects, however what made this an outlier case was the Chinese loans provided for developing oil and mineral extraction infrastructure (Niambi, 2018). In any case, these two outlier cases did not distort the analysis too much because the robust regression method was used.

Second, this is important for selecting the right case for the process tracing analysis. As can be seen from the figure, Cameroon lies on the regression line. Furthermore, the country has received substantial lending from China relative to its GDP. Therefore, the case of Cameroon fits the criteria for the process tracing case laid out in the previous chapter. In this country, a typical relationship between Chinese lending and economic liberalisation is indicated by the regression and the significant Chinese lending means that the process of how Chinese lending leads to less neoliberal policymaking should exist and can be traced. Thus, Cameroon is the case that was analysed using the process tracing method. In the next subsection, the analysis of the direct way that Chinese lending leads to less neoliberal policymaking will be presented. H2a-H2b will be either confirmed or disconfirmed.

5.2. Process Tracing Results

In this section of the chapter, the process tracing analysis results will be presented. As was mentioned in the previous section, the selected case for the process tracing analysis is Cameroon, as it is a typical case according to the linear regression analysis. Two causal mechanism paths were analysed, the path where Chinese lending would directly affect policymaking in developing countries (marked by hypotheses H2a and H2b), and the path where Chinese lending would indirectly affect policymaking in developing countries by causing IFIs to make their conditionality less stringent (marked by hypotheses H3a and H3b). The results for the analyses of each of these paths will be presented in the next two subsections.

5.2.1. Direct Path

The first causal mechanism suggests that developing countries that receive Chinese lending should become less reliant on IFIs. This means that Cameroon should replace a part of their IFI reform-based lending with Chinese lending, which should be demonstrated by lesser amounts borrowed from IFI policy reform-based programmes in the years when they borrow more from China. This is stated by H2a. Participating in less policy reform-based lending programmes should result in less neoliberal conditions accepted, and, in turn, less neoliberal policies enacted. However, if Chinese lending were to actually lead to less neoliberal policymaking, the projects that are funded by loans from China should be accompanied with less neoliberal reforms than similar projects financed by IFIs. This is what H2b states.

(29)

29 5.2.1.1. Changes In Borrowing From IFIs

Figure 3.

Figure 3 displays how IFI lending and Chinese lending per year to Cameroon changed during since the 2000s, as well as how much IFIs lent globally. Cameroon borrowed 4 million USD from China in 2003 and 5 million USD in 2006, however more significant Chinese lending began in 2007, when Cameroon borrowed 67 million USD. Chinese lending skyrocketed in 2011-2012, when Cameroon borrowed 583 and 1522 million USD respectively. Since then, China lent at least 400 million USD to Cameroon every year.

IFI lending to Cameroon fluctuated between 20 and 143 million USD each year before Cameroon began borrowing from China and started fluctuating at a higher 30 to 244 million USD range between 2007 and 2016. At a first glance, it seems that Chinese lending was not a substitute for IFI lending but rather a supplementary source of financing. Not only did IFI lending not decrease after Cameroon started to borrow from China but average borrowing from IFIs has increased. Furthermore, the most significant amount of IFI lending was in 2009 and 2012, the same years as China massively increased its financing of projects in Cameroon.

However, it is important to put these trends into context. First, IFI lending globally was significantly lower before 2008, after which it increased more than twofold. Therefore, if Cameroon was to follow global trends, a significant jump in borrowing amounts would have happened instead of the slight increase. It is possible that IFI lending to Cameroon did not increase as much as their average lending to all developing countries because a part of Cameroon's financing needs were met by Chinese loans.

The effects of Chinese lending on the economic policymaking in developing countries : does an access to Chinese loans lead to less neoliberal policies?

Master Thesis Political Science

Specialisation Track Political Economy