Introduction

As Malaysia’s largest trade (and increasingly investment) partner,Footnote 1 China is widely seen as an avenue by which Malaysian economic fortunes can be revitalized. This viewpoint is understandable as gross domestic product (GDP) growth has slowed down noticeably since the late 1990s, raising concerns that Malaysia might be locked-in a middle-income trap (Cherif and Hasanof 2019). The sentiment has also seemingly been reinforced since the Belt and Road Initiative (BRI)—arguably China’s most important engagement tool with the rest of the world—was announced by Chinese President ** in 2013. Indeed, successive Malaysian political leaders have sought to leverage opportunities brought about by the BRI. In particular, former Prime Minister Najib Razak (2009–2018) saw it fit to foster closer ties with the Chinese economy. Reflecting the increasing relevance of the BRI in China-Malaysia relations, Malaysian analysts have been quick to explore common ground with the Chinese on the five types of connectivity (finance, trade, infrastructure, policy, and people-to-people) promoted by the initiative (see, for example, Liu and Lim 2019; Zhang et al. 2017; Ngu and Ngeow 2021; Kong 2016).

With the above as a backdrop, this paper seeks to understand how connectivity has been forged between both countries. To this end, it sheds light on the connection between Chinese economic engagement and policy experiments by key Malaysian stakeholders, not least the political elites and bureaucracy. It focuses primarily on financial connectivity, especially the relationship between Chinese foreign direct investment (FDI) entering Malaysia and the structural transformation that results. More specifically, the paper analyses how Malaysian industrial upgrading efforts are interacting with increased financial (and by extension, trade) integration with China. Along the way, it will examine how Malaysian policymakers—particularly since the 1990s—have made sense of and engaged with economic opportunities opened up by the Chinese market. In addition to the lock-in impact of these moves on industrial deepening, the paper will explore how recent waves of Chinese FDI (under the banner of the BRI) have seemingly been less than carefully incorporated into the domestic setting.

The paper sheds light on structural transformation in the Malaysian economy by unveiling potentially unobservable mechanisms linking economic data at the macro-level with meso-scale empirical analysis in the real estate industry (a key component of the services sector). Two inter-related arguments are forwarded. First, by opting for a closer economic integration with China, Malaysia’s industrial progress has been truncated, witnessed especially in its lower emphasis on manufacturing and ‘proper’ industrialization. Second, the paper argues that Chinese FDI entering Malaysia is composed primarily of services. The services-heavy nature of Chinese FDI brings about a low economic multiplier effect, while exacerbating societal inequality and political tension. While Malaysia’s deindustrialization predates its moves to establish closer economic ties with China, welcoming these types of Chinese FDI seems to have complicated its already tenuous upgrading pathway. More specifically, the concentration of such capital in a critical economic corridor (e.g. Iskandar Malaysia) has locked-in an ecosystem that already underinvests in long-term capability building.

This paper adopts a mixed method approach, drawing conclusions based on a close examination of several economic databases and fieldwork findings. For the former, relevant statistics were extracted from the United Nations Commodity Trade Statistics Database (UN Comtrade) and United Nations Industrial Development Organization (UNIDO). Statistical information was also retrieved from the Department of Statistics Malaysia (DOSM), especially those related to the stock and flow of FDI in the country. In instances where publicly available data was not available, the authors sought the assistance of key DOSM officials for more information. For the latter, the authors interviewed policymakers, business executives, and academic and think tank researchers, who have been involved in China’s economic engagement with Malaysia. These interviews were conducted in Kuala Lumpur, Iskandar Malaysia, and Singapore from 2016 to 2018, with the goal of unearthing the broader political dynamic sha** China–Malaysia economic relations. To improve the reliability of the interview data, it was cross-validated and triangulated with other relevant archival and secondary materials such as annual reports of major companies, government reports (especially those by the Malaysian Investment Development Authority), newspaper reports, and technical manuscripts. Given the sensitive nature of the issues discussed (such as business-state interactions), all interviewees were promised confidentiality.

The paper is organized in the following manner. The next section reviews literature on China ‘Going Out’, critiquing its lack of engagement with industrial upgrading in the host economies. The third section provides a wider context by analysing the transformation of the Malaysian economy since its 1957 independence. Particular attention is paid to events since the 1990s. Subsequently, it analyses how closer economic ties between China and Malaysia, particularly in trade and investment, are linked to premature deindustrialization. It explores various descriptive statistics and casts light on the impact of Chinese FDI in Iskandar Malaysia, the Southeast Asian nation’s most prominent economic corridor. The goal is to unpack how Chinese FDI has proscribed the corridor’s development trajectory, in addition to exacerbating socio-political discontent. The paper then discusses the core findings and situates them within the existing scholarship. The final section concludes and offers some policy implications.

Present State of Affairs: A Critical Review

Scholarship on China ‘Going Out’ has—until relatively recently—focused on the agency and perspectives of Chinese stakeholders. In the early stages, the emphasis was almost exclusively on the Chinese state (e.g. Arase 2015; Beeson 2016). Subsequent works have pushed the paradigm further, detailing the need to analyse other stakeholders from both within China and without. For example, Liou (2014) argues that Chinese engineering state-owned enterprises (SOEs) have internationalized their operations under the pretext of the BRI, but their actions do not always reflect national interest. In some cases, they have even gone against Bei**g’s directive (see Liou 2009). Another perspective is offered by Hofman (2016), who demonstrate that, in the Tajikistani agriculture industry, small-scale Chinese family farms are the main drivers of the BRI. Unlike the more cumbersome and bureaucratic SOEs, these family farms are primarily motivated by profits and have since exploited previously unheralded market niches in the Central Asian economy (see also Bräutigam and Tang 2012; Lee 2017).

Enriching the perspective are studies that scrutinize stakeholders in the host economies and their interest. In his analysis of Chinese-financed infrastructure in key Southeast Asian economies, Camba (2020) argues that these projects hinge upon the formation of a coalition (from within and outside the host) that combines political, economic, and organizational resources from the relevant actors. Nevertheless, the coalition is subject to change and could be fractured when domestic dissenting forces (at both the central and subnational levels) reach a critical mass. When this happens, the projects suffer delays (or worse, cancellation) (see also Calabrese and Cao 2021).

Notwithstanding its insights, the scholarship above has generally paid scant attention to issues pertaining to industrial upgrading (cf. Lin and Wang 2017). This is odd given that the BRI (and more broadly, Chinese capital exports) is supposed to be a ‘win–win’ proposition. It is even more perplexing as a large part of Chinese capital exports have gone towards fellow develo** countries and increasingly more host countries are devoting specific desks/teams to attract Chinese firms (see Dharma and Suryadinata 2021). For example, Lin (2012) argues that a cohort of large middle-income countries, spearheaded by China, is increasingly resha** South-South cooperation. Almost 100 million labour-intensive manufacturing jobs is estimated to be shed from these economies to other develo** economies further down the pecking order. Lin postulates that the onus falls on those develo** economies that have the political will and bureaucratic nous to implement strategies to embed FDI from China (and the rest of the large middle-income economies). Injecting nuance to this perspective are global value chain scholars, who stress that firms in develo** countries can promote ‘strategic coupling’ with their counterparts from China (Yang 2016; Yeung 2009). According to Yang (2016), the spatial expansion of production networks (especially those in the coastal area) to certain Southeast Asian economies could more purposefully incorporate firms from the latter into Chinese-led production networks, stimulating intra-regional cooperation and local development.

However, recent research has dampened expectations. According to Tham et al. (2016), Southeast Asia’s growing integration with China has been correlated with minimal product upgrading in key commodities such as information and communications technology (ICT) goods. Their analysis dovetails with Coxhead (2007), who warn that China’s rapid growth is detrimental to the Southeast Asian economies. His thesis is centred on China’s voracious demand for natural resource products, which in turn alters world market prices and potentially traps Southeast Asia’s smaller economies in a ‘new resource curse’ (cf. Zhang et al. 2017; Gomez et al. 2020). Relatedly, Lim (2019) demonstrates that Chinese FDI entering Southeast Asia is unlikely to stimulate upgrading and foster linkages, at least not to the extent offered by the region’s more established investors, particularly Japan. His rationale hinges on the nature of Chinese FDI—much has been directed towards services such as real estate development and wholesale and retail trade. Real estate poses a particular conundrum to the host economies, especially when it finances luxury enclaves offering little positive spillover to the local stakeholders. Contrarily, Japanese FDI entering Southeast Asia has, since the 1990s, been largely financing secondary industries (i.e. manufacturing).

The services-heavy nature of Chinese FDI resonates with literature on deindustrialization. Scholars such as Palma (2005) and Rodrik (2016) highlight a significant deindustrialization trend in recent decades. Some of the most obvious indicators include loss of manufacturing output and employment as well as the growing weight of services in the economy. What is more, this phenomenon has taken root not only in advanced economies, but also in develo** ones. According to Rasiah (2011), deindustrialization must be understood contextually. It is considered positive if manufacturing's productivity and trade performance remains strong, but negative if they decline. His thesis builds on Rowthorn and Wells’ (1987) analysis of the US and UK economies, in which the former is said to experience positive deindustrialization while the latter negative deindustrialization. Implicit in such analysis is the possibility of services-led growth. For example, Tregenna (2011) argues for a closer examination on the technological-organisational and sectoral characteristics of the economic activities in question, advocating a more nuanced understanding of deindustrialization.

Yet, the reality is that services, relative to manufacturing, tend not to be very dynamic and internationally tradable. In other words, it is no longer apparent that manufacturing offers opportunities to aspiring develo** countries eager to promote structural transformation, at least not to the same degree it did to the Global North (see Chang 2012; Haraguchi et al. 2017). There are exceptions—certain service activities are highly productive and tradeable across borders. However, the catch is that these services are typically highly skill-intensive and do not possess the scale to absorb labour, at least compared to manufacturing. The twin realities do not augur well for develo** economies that have low- to medium-skilled labour in abundance (Rodrik 2016).

As alluded to earlier on, deindustrialization (or at least, some of its symptoms) in several Southeast Asian markets has seemingly occurred following closer ties with the Chinese economy. Gallagher and Porzecanski (2010) observe similar trends in Latin America and the Caribbean (LAC). However, they also emphasize that the relatively low levels of industrialization of the LAC economies could, to a large extent, be attributed to their adoption of Washington Consensus-inspired economic ideas since the 1980s. The liberalization of trade and investment, in particular, opened up opportunities for more competitive (Chinese) products to flood their markets, indirectly stunting the growth of local producers.

Notwithstanding such insights, it must be noted that these works mainly utilize trade data. There is thus a need to also investigate FDI dynamics, as more contemporaneous research has shown. For example, in a recent study of Brazil’s economic relationship with China, Hiratuka (2022) argues for a more rigorous analysis of Chinese FDI entering the Brazilian economy, in addition to conventional trade analysis. Focusing on the investment activities of BYD, one of China’s top players in electric vehicles, he delineates important possibilities for new economic relations with China that go beyond trade in commodities, not least productive FDI that can potentially help Brazil simultaneously incorporate more knowledge-intensive activities and drive positive environmental impacts by reducing carbon footprint. However, the transformation of these possibilities into effective benefits requires coordinated actions on the part of Brazilian policymakers. This Brazil-specific experience resonates with more mainstream LAC industrial policy scholarship. Peres and Primi’s (2019) critique of some relatively successful industrial upgrading cases share multiple commonalities. Relative to the underperformers, those LAC economies that performed better displayed clearer policy priorities, fostered more constructive dialogue between the public and private sectors, and more effectively mobilized investment in bundles in critical areas (e.g. infrastructure, skills, and finance).

To build on and enrich the above body of knowledge, especially within the context of Southeast Asia, it is important to unpack how Chinese FDI has influenced (de)industrialization in Malaysia. A detailed reading of how Malaysia’s upgrading trajectory has been shaped by Chinese-financed activities is necessary to better understand the process interconnecting the BRI and host state structural transformation. In addition to fostering a closer dialogue between international economic forces and domestic industrial development, the paper illustrates how capital flows and their internalization process complicates long-term growth. Of particular interest is how certain types of Chinese FDI are entrenching the domestic milieu into services that are arguably counterproductive to expedite technological progress. These dynamics will be teased out in the subsequent sections.

Structural Transformation in the Malaysian Economy

Structural transformation in the Malaysian economy follows a common pattern experienced by most develo** countries—a shift from agriculture to the industrial sector, followed by an expansion in the services sector. The country has moved from a resource-based economy focusing on rubber and tin, to the manufacturing of consumer goods (in the early 1960s), and subsequently to semi-skilled and automation-oriented industries in the 1970s and 1980s. Figure 1 shows two key structural transformation junctures in the Malaysian economy. The first structural shift occurred between the agriculture and the manufacturing sectors in the late 1980s. The share of the manufacturing sector overtook agriculture in 1987 and further increased threefold to its peak in the year 1999 (30% of GDP). On the other hand, the agriculture sector has shrunk to less than half of its size in the 1970s and currently only occupies less than 10% of the country’s GDP.

Fig. 1
figure 1

Source World Development Indicators (WDI Database)

Structural transformation in the Malaysian economy, 1971–2020 (% of Gross domestic product and % of manufacturing value-added).

Full size image

The second shift occurred in the early to mid-2000s where the divergence between the services and manufacturing sectors became prevalent. More specifically, the share of the manufacturing value-added started to decline while that of services witnessed a gradual rise circa 2006. The decline of manufacturing per se is not overly worrying. However, it becomes a concern when the manufacturing value-added of the medium and high-technology industries has also tapered off at about the same time. Such development suggests the emergence of a middle-income trap where the country has attained a certain level of per capita income on the basis of labour-intensive activities (usually in mature industries like agriculture), but are struggling to move towards more sophisticated activities that yield more value (Cherif and Hasanof 2019).

A similar perspective is reflected in the Southeast Asian economy’s employment trends. Figure 2 illustrates that employment in the agriculture sector has fallen from 31 of total employment in 1982 to 9.5% in 2020. While services continued their upward trend, manufacturing overtook agriculture as the employment generation engine in the 1990s. Figure 2 also implies that excess labour from the agriculture sector has been slowly absorbed into manufacturing, given the steep increase in employment in manufacturing (before the turning point).Footnote 2 This contributed to the rise in employment and wage increase in the manufacturing sector. However, the uptrend momentum started to plateau in subsequent years as the share of manufacturing employment stagnated for almost a decade from 1992 to 2000, before commencing a slow decline for more than 15 years from a share of 23% to 16.7% in 2020.Footnote 3 The contraction of employment within the manufacturing sector may be due to the shift into automation that replaces unskilled workers especially in the electrical and electronic industry, and some food and packaging industries (Malaysia Productivity Corporation 2017). The 2016 Economic Census shows that employees in the manufacturing sector are overwhelmingly concentrated in the plant and machine operators, and assemblers category, in which required skills are semi- to low-skilled levels—and thus easily replaceable by machines and automation (Kang and Kam 2019). This, however, should not be interpreted that there are no skills-oriented activities within the Malaysian manufacturing sector. It merely implies that the high-skills and high-value manufacturing activities do not have the critical mass to drive the manufacturing sector at large.

Fig. 2
figure 2

Source Department of Statistics Malaysia

Employment trends in key sectors, 1982–2020 (% of Total employment).

Full size image

While it is not within this paper’s remit to explore in substantial depth policies that led to the decline of manufacturing (and the rise of services), existing studies have offered a few explanations contributing to this development. The lowest common denominator of these studies points to the state’s role in in the promotion of ineffectual industrial policies (see, for example, Jomo 2007; Gomez et al. 2021; Rasiah 2011). First, there appeared to be an overreliance on FDI to drive technological progress, especially in high-technology manufacturing. While this per se is not a huge issue in the 1970s-1990s, the strategy has been strained since the late 1990s. Coupled with Malaysia reaching its Lewisian turning point is the emergence of a newer group of develo** countries promoting pro-business policies and enjoying even lower labour cost (relative to that of Malaysia), which in turn heightens competition to attract the same pool of investment dollars. Within ASEAN itself, Malaysia started to face intense competition from the Cambodia, Laos, Myanmar, and Vietnam (CLMV) economies from this period onward (see Amakawa 2010; Fujita 2013). In the immediate peripheries of ASEAN, the Chinese and Indian economies were also coming into their own as large, young economies open to transnational corporations (TNCs).

Relatedly, Malaysia’s unorthodox response during the 1997 Asian financial crisis, which ravaged major regional economies, has brought about a mixed response from the international community. In the beginning, the policy measures, in particular the imposition of capital controls, were perceived sceptically. Some deem the moves as a step toward decoupling the country from international financial markets, although more detailed analysis suggest that they helped in stabilizing market confidence during the peak and latter stages of the crisis (Sheng 2008; Singh 2022). In any case, investor confidence only returned by the mid-2000s, although by which time they were awash with options in the form of other low-cost economies such as the CLMV (Menon 2014).

Third, policy attention during the mid-2000s shifted to the promotion of regional economic corridors, which cut across various Malaysian subnational states. The idea was to exploit the spatial spread of synergies more purposefully by supporting lead sectors, especially in the less developed areas of the country. In 2006, the first corridor—Iskandar Malaysia—was launched in the southern tip of West Malaysia. By the early 2010s, four other regional corridors were established. Iskandar Malaysia still ranks as, by far, the most prominent corridor, based on various economic indicators (Rasiah and Krishnan 2020). Analysts point out two main issues with the initiative. For one, the corridors displaced subnational governments as the main stakeholders responsible for development within their jurisdiction as a cohort of officers were established and parked under the powerful Prime Minister Office (Hutchinson 2015). In addition to sparking off federal-subnational bureaucratic tussles, the industries targeted in some of the corridors became overly diversified. This is especially salient in the case of Iskandar Malaysia. Van Grunsven and Hutchinson (2016) depict that while manufacturing activities, namely electrical and electronics (E&E), petrochemical and oil and gas, as well as food and agro-processing are mentioned in official documents, six new services have been assigned priority. They are tourism, education, creative industries, global business service, healthcare, and logistics. Extra incentives are offered to prospective investors in these six services, but not for manufacturing (IRDA 2021). The apparent slant towards services was more of a case of policymakers’ view of manufacturing. In particular, the E&E sector was ‘seen to… be “unstable”, and insufficiently value-added, with the implication that marketing and promotion efforts should be directed elsewhere’ (van Grunsven and Hutchinson 2016, p. 83).

Empirical Evidence

China-Malaysia Economic Relations: Recent History

Diplomatic relations between Malaysia and China were formally etched in May 1974 when then Malaysian Prime Minister Abdul Razak Hussein (1970–1976) visited Bei**g and established diplomatic ties with his Chinese counterparts. While this first point of contact was ground-breaking, it was not until the 1990s that the relationship flourished. Led by Mahathir Mohamad, Malaysia’s longest-serving Prime Minister (1981–2003; 2018–2020), the Southeast Asian nation began to see China as an economic partner that presented more opportunities than threats (Khoo 2021). Further integration into Malaysia’s economic network was initiated when Mahathir invited China to attend the 1991 ASEAN meeting in Kuala Lumpur.

In 1997, the trust between the countries was further elevated when Mahathir proposed a new regional pact—the East Asia Economic Caucus (EAEC)—that simultaneously encompassed most of East Asia and excluded Western nations. Although the EAEC was not successful, it was viewed as an important gesture to include China in Malaysian (and Southeast Asian) economic endeavours. Chinese support during the 1997 Asian financial crisis, primarily the open pledge that it will not devaluate the Chinese Yuan (CNY), bolstered China–Malaysia ties even more (see Khor et al. 2022). Notwithstanding the broad dimensions of cooperation between China and Malaysia (as highlighted in multiple joint statements),Footnote 4 a running theme in all these joint statements is a focus on economic fundamentals. Common areas of cooperation include the exploitation of comparative advantages between respective economies to strengthen and deepen economic cooperation, the reduction of tariff and non-trade barriers, and a call to improve cooperation between the two countries in the services sector such as transportation, education, energy, and finance.

Malaysia’s economic engagement with China was also enhanced through ASEAN with the signing of The Framework Agreement on Comprehensive Economic Cooperation in 2002. The framework formed the legal basis for further integration between Malaysia and China through the creation of the ASEAN-China Free Trade Area (ACFTA). Under the ACFTA three key agreements were signed starting with the ASEAN-China Free Trade Agreement (ACFTA) in Goods (29 November 2004), followed by the ASEAN-China Trade in Services Agreement (14 January 2007), and finally ASEAN-China Investment Agreement (15 August 2009). The chronological process of the creation of ACFTA is shown in Table 1. The main objective of the agreement in goods was for tariff reduction and elimination of tariff lines categorized under either ‘normal track’ or ‘sensitive track’. The services agreement aimed to liberalize and eliminate discriminatory measures concerning trade in services, and the investment agreement seeks to promote and facilitate investment flows.

Table 1 Agreements signed between ASEAN and China in the formation of ASEAN-China Free Trade Area.
Full size table

The ACFTA has created a boom in Malaysia’s exports and imports to/from China. Figures 3 and 4 respectively illustrate Malaysia’s trade with its key partners. Compared to other traditional foreign markets, both figures show that after 2002, Malaysia’s exports and imports with China have increased significantly. Several reasons undergird this increase. For one, there is the stage-by-stage tariff reduction brought about by the ACFTA. China also officially joined as a member of the World Trade Organization (WTO) in late 2001, which helped it enhance economic ties with other economies such as Malaysia (Tham et al. 2016). Exports to China are mainly driven by manufactured products, primarily electrical and electronics (E&E) goods, chemical products, optical and scientific equipment, and other resource-based products such as liquefied natural gas and rubber products. The US, which was once Malaysia’s top export destination from 1998 to 2006, was replaced by China from 2009. China eventually became the largest exports destination in 2019, overtaking neighbouring Singapore. As for imports, China’s dominance was felt more palpably when it took the number one spot from Singapore in 2012. Since then, imports from China have held a commanding lead over those of the other traditional economies. Reflecting an increasingly tighter production network, Malaysia mainly brings in E&E components, machinery, and equipment products from China (see also Tham and Kam 2014). Overall, closer trade ties with China have induced both Malaysian as well as foreign-invested companies to alter their investment and production patterns within the country as well as across the region, although this impact is of an indirect nature. For example, in the tabling of Budget 2020, a ‘China Special Channel’ was established to capitalize on the possibility of trade and investment diversion caused by the increasingly tense US-China disputes (PwC 2019). More specifically, it seeks to attract high value-added, high impact activities from the ongoing exodus of China-based TNCs seeking to mitigate their sourcing risk, in addition to those planning to establish a presence in East Asia.

Fig. 3
figure 3

Source UN Comtrade database

Malaysia key export markets, 1989–2020 (USD Billions).

Full size image
Fig. 4
figure 4

Source UN Comtrade database

Malaysia key import markets, 1989–2020 (USD Billions).

Full size image

Figure 5 shows the net FDI flows into Malaysia from 2011 to 2020. Net FDI flows from China started to pick up from 2013 (the year the BRI was announced) before peaking in 2017. This surge meant that Chinese FDI was able to challenge the other traditional FDI providers such as Japan and the US. However, this momentum was cut short when it suffered a decline in 2018. Chinese net FDI flows into Malaysia recovered modestly in 2020, which saw China climbing to the second spot. The rather volatile nature of Chinese FDI flows can be contrasted with those originating from Japan (and to a smaller extent, Singapore). Although this paper does not explore in depth the reasons behind Chinese and Japanese FDI flow patterns, a likely explanation is the heavy presence of manufacturing firms driving the latter. Lim (2019) estimates that more than 42% of Japanese FDI entering Southeast Asian economies have financed manufacturing activities between 2012 and 2017. Only an average of 11% of Chinese FDI have gone into manufacturing during the same period. The broader argument, highlighted in the earlier sections, is that manufacturing—courtesy of its value chain linkages to the domestic industrial ecosystem—is bound to generate more long-lasting and widespread impacts than investment related to other sectors (especially services) (Rodrik 2016).

Fig. 5
figure 5

Source Statistics of Foreign Direct Investment in Malaysia, Department of Statistics Malaysia (converted into USD)

Net foreign direct investment flows entering Malaysia, 2011–2020 (USD Billions). *Negative value indicate an outflow or a decrease in investment.

Full size image

A more revealing picture can be had when the sectorial breakdown of Chinese FDI is undertaken. Figure 6 illustrates that, based on net FDI flows, Chinese investors have invested mostly in services, followed by the manufacturing sector. Indeed, net FDI flows entering the services sector occupies the top position in three out of the five years observed. Nevertheless, net FDI flows have declined sharply from 2017 to 2018, with flows in services declining more than those of manufacturing. Perhaps more noteworthy, FDI flows financing the services sector recovered earlier than those of manufacturing from 2018 to 2020. Further data provided by DOSM on China’s FDI stock reveals a broadly similar story. From 2016 to 2020, the spread between FDI stock in the services sector relative to the manufacturing sector amounted to about USD1 billion almost every year.

Fig. 6
figure 6

Source Statistics of Foreign Direct Investment in Malaysia, Department of Statistics Malaysia (converted into USD)

Net foreign direct investment flows from China, 2016–2020 (USD Billions) (DOSM classifies agriculture and construction activities under ‘other sectors’ in this particular set of data.).

Full size image

To add more context, Table 2 details the breakdown of the total capital invested within the services sector. Although the table does not separate domestic investment from foreign-sourced ones, they still shed useful insights on the services sector. For one, real estate is the most important player, contributing about 38% of the sector between 2015 and 2020. While it is true that real estate has sparked some positive externalities (e.g. employment opportunities), Malaysia’s current income level requires structural transformation that is of a higher order, as discussed in the earlier sections.Footnote 5 Additionally, it is not immediately apparent that real estate is highly productive and tradeable across borders. If anything, some of the most capital-intensive real estate undertakings (promoted by certain Chinese TNCs) might actually be antithetical to Malaysia’s current phase of industrialization.

Table 2 Total capital invested by domestic and foreign investors in the services sector, 2015–2020 (USD Billions).
Full size table

Of Manufacturing and Services: Macro-level Trends

This paper further explores industrial upgrading by analysing simultaneously the trade in domestic value-added (DVA) and bilateral revealed comparative advantage (RCA) indicators, as has been utilized in Kam (2017).Footnote 6 The justification is as follows: an increase in DVA content in export implies an increase in local economic activities. If these local activities are of high quality or value, then they are likely competitive in the international arena. In other words, a high DVA and RCA imply that domestic producers have local advantages (such as labour, capital, and trade policy regimes) to specialise in the export of a particular product compared to other nations, industries, and/or companies (Kam 2017).

A caveat to this interpretation, however, is that Chinese FDI involvement in these exports may not be fully reflected in the DVA and RCA indexes. For example, an increase in DVA content in export to China is not necessarily initiated by Chinese firms that have invested their facilities in the FDI recipient state (in this case, Malaysia). The increase could have been sparked by TNCs originating from other economies and/or local firms that have export interest with China. Therefore, upgrading within the host economy might not directly be attributed to Chinese FDI (vertical spillover). However, exposure to the Chinese market is likely to create an indirect impact (i.e. horizontal spillover) whereby local firms are incentivized to upgrade to stay competitive in the Chinese market. Put simply, regardless of whether there is direct involvement from Chinese FDI or not, the need to stay competitive in the Chinese market requires local establishments to adapt and upgrade their activities. This applies to both manufacturing, and services exports. Working with these assumptions, this paper proxies the tandem movement between DVA and RCA as a form of local upgrading. In principle, an increase in DVA content in Malaysian exports to China implies that local firms have technologically upgraded and can replace previously imported inputs. Additionally, to ensure this increase in DVA represents an industrial upgrade, the goods produced should become (or remain) competitive in the Chinese market.

Figure 7 shows the DVA content of manufactured export and the related bilateral RCA between 1995 and 2018. From 1995 to 2004, the DVA share of Malaysian export to China in manufacturing has been decreasing, implying that more imported inputs have been used to generate exports to China. Yet, since 2004, there has been more domestic content in export to China, while the products exported show an increasing comparative advantage. These trends suggest an increase in local capabilities in manufactured export targeting the Chinese market. However, there is also a noticeable drop in the competitiveness of Malaysian manufactured export to China after 2014, despite an increase in DVA content. Several interpretations can be forwarded. For one, even though Chinese FDI has increased, the local activities brought about by these investors may not be high value-added. This drop in RCA of manufactured export might also imply that Chinese FDI have largely financed non-manufacturing (usually services) activities and/or underfinanced sophisticated manufacturing activities. A more pessimistic conjecture is the correlation of Chinese FDI with entrenchment in low value-added activities.

Fig. 7
figure 7

Source OECD Trade in Value-Added (TiVA) database

Domestic value-added content of manufactured export and bilateral revealed comparative advantage of manufactured export, 1995–2018.

Full size image

Figure 8 shows that the DVA content of services export to China has been increasing. Although services export is deemed ‘uncompetitive’ (RCA value below one) throughout most of the period analysed, the index is still increasing in nature. In 2014, when there is a decline in manufacturing export competitiveness (see Fig. 7), the competitiveness of services exports to China increased significantly, moving very close to the ‘competitive’ category (RCA value above one). The upswing in competitiveness also occurred in tandem with the increase in DVA, implying some upgrading in the services sector (see also Kam 2017). Nevertheless, the competitiveness of services exports fizzled in subsequent years.

Fig. 8
figure 8

Source OECD Trade in Value-Added (TiVA) database

Domestic value-added content of services export and bilateral revealed comparative advantage of services export, 1995–2018.

Full size image

Put together, the two figures illustrate that the DVA of the services sector has been increasing gradually since 1995, while that of manufactured export hovered around the 0.5-level throughout the period analysed. Compared to services export, Malaysian manufactured export has been generally competitive in China. However, both their competitiveness has declined somewhat after 2014. Between 1995 and 2018, domestic activities in services have grown in tandem with competitiveness in the Chinese market. This relationship is less clear cut for manufactured export, although its RCA declined amidst an increase in DVA post-2010. The prognosis thus far underlines the cree** weightage of Chinese FDI in services, which indirectly alters the incentive structure for Malaysian firms and truncates long-term manufacturing performance. While this assessment is derived from macro-level data analysis, it is supported by further empirical research in Iskandar Malaysia, detailed in the next section.

Chinese Investment in Iskandar Malaysia: Meso-Scale Observations

How has the services-heavy nature of Chinese FDI interacted with the development trajectory of the Malaysian ecosystem then? Some insights can be inferred from the aforementioned Iskandar Malaysia, the country’s oldest and most important economic corridor bordering wealthy Singapore. This corridor came about in 2006 as the Malaysian federal government sought to stimulate growth in the country’s southernmost state of Johor. Launched as one of the high-impact projects under the Ninth Malaysia Plan (2006–2010), this corridor covers an area as wide as 9300-acre (Khazanah 2006). Liberalizing most economic activities and offering generous fiscal incentives, the plan was for Iskandar Malaysia to capture the spill-over effects from Singapore, akin to the relationship between Shenzhen and Hong Kong (Rizzo and Glasson 2012). However, Iskandar Malaysia’s growth has generally been uneven since its inception (Elisabetta 2019; Hutchinson 2015). This can be seen particularly in the launching of luxury homes—the numbers have consistently risen since 2010. In spite of demands for low- and medium-cost houses, housing supply has been skewed towards higher-end units. Research also demonstrates that private housing has clearly exceeded the housing affordability cut off point (i.e., median housing price) of MYR260,000 (USD68,000) (see Ng and Lim 2017).Footnote 7 According to reports, real estate investment accounted for MYR35.1 billion (USD9.2 billion), constituting 33.1% of the total investment capital in the region from 2006 to 2012 (IRDA 2014, 2016). This in turn suggests a heavy presence of services as real estate accounts for close to 38% of the entire sector’s capital (see Table 2).

The bias towards real estate (and services) deepened following the 2013 launching of the BRI. As the Chinese economy decelerated following years of rapid growth, its cohort of real estate TNCs targeted overseas markets like Iskandar Malaysia, diversifying their hitherto China-centric portfolio. Country Garden (one of China’s largest TNCs) is the biggest player in Iskandar Malaysia, sponsoring three projects. Forest City by far ranks as the firm’s most consequential undertaking due to its sheer size. It is also the most capital-intensive project in Iskandar Malaysia. The mushrooming of these Chinese-financed projects, nevertheless, worsened Iskandar Malaysia’s mismatch of homes, hitting the ethnic Malays particularly hard (Ng and Lim 2017).Footnote 8

Forest City illustrates these dynamics well. Valued at a total investment of MYR220 billion (USD58.0 billion), the project is driven by the aforementioned Country Garden partnering Esplanade Danga 88 Sdn Bhd, a firm affiliated with the Sultan of Johor (Aw 2014). Perhaps more importantly, it ranks as arguably the costliest Chinese project in Malaysia. While there are of course other high profile Chinese-financed undertakings, ranging from the East Coast Rail Link (ECRL) to Malaysia-China Kuantan Industrial Park (MCKIP), their expected financial commitment (equity, debt, and/or a combination of both) do not come close to that brought in by Forest City (see Gomez et al. 2020; Lim et al. 2022).Footnote 9

While still at an early phase of development, Forest City is to eventually encompass four manmade islands. Spread over 1386 ha, these islands are expected to house 700,000 people.Footnote 10 For Country Garden, it aggressively marketed the project to a predominantly overseas (mainly Chinese) clientele. Through its taglines, ‘duty-free island-based living next to Singapore’ and a ‘special economic zone within a special economic zone’, Forest City is marketed to parties interested in acquiring a second-home or retirement property (Interview, Johor Bahru, 7 May 2017). Indeed, the Chinese TNC brought in planeloads of prospective buyers from China (especially the growing middle class), giving them tours of the project. During these tours, Country Garden employees conveyed the message that Forest City and the surrounding area are anticipated to experience rapid growth in the near future (Mahrotri and Choong 2016).

In their defence, Country Garden executives stress that Forest City offers not only luxury homes. It also promotes industrial development by attracting businesses to operate eight key sectors: tourism and meetings, incentives, conferences and exhibitions (MICE); healthcare; education and training; regional headquarters; offshore finance; e-commerce; emerging technology; and green and smart industry (see Wong 2016). However, a closer look at these eight sectors reveal that they are almost entirely services in nature. Furthermore, nearly 70–80% of Forest City is designated for high-end residences, with the remaining 20–30% allocated for industrial development. This spatial arrangement in turn caps how much services (and to the extent that they are highly productive and tradeable) could be embedded into the project. This disjuncture was raised during the authors’ fieldwork, although there were no credible counterarguments offered (Interview, Johor Bahru, 7 May 2017; Interview, Singapore, 24 December 2017). Regarding the lack of specific attention to attract manufacturers, the feedback gathered was that Country Garden executives are concerned that manufacturing firms could end up consuming too much space. They are also worried that these companies could discharge environmentally damaging effluent, subsequently jeopardizing the supposedly pristine environment that the project claims to offer (Telephone interview, [Singapore to] Johor Bahru, 9 May 2018).

Parallel to the above insight is the executives’ lack of understanding of the broader policy environment of Iskandar Malaysia. One of them was extremely surprised to discover that the corridor, prior to Forest City’s introduction, has not attracted as much manufacturing activities as he/she initially thought (Telephone interview, [Singapore to] Johor Bahru, 9 May 2018). This individual further speculated that one possible reason could be due to Country Garden’s hiring of McKinsey, one of the world’s best-known management consultants, as its strategic planner for the project. The reliance on such a renowned firm might have led to insufficient in-house research. While this paper does not focus specifically on the business strategy of Country Garden in Malaysia, the Chinese TNC’s general lack of local knowledge has been flagged in other reports (e.g. Liu and Lim 2019; Koh 2021).

Notwithstanding the above, Country Garden’s industrial promotion personnel were generally more aware of international business trends and multilingual compared to their colleagues marketing residential units (Interview, Johor Bahru, 7 May 2017; Interview, Singapore, 24 December 2017). For the latter, it was noticeable that homebuyers from China, relative to the rest of the world, were actively targeted. This preference is perhaps due to Country Garden’s lack of exposure outside China, evidenced in how the majority of the residential marketing team could only converse in the Standard Chinese (Mandarin) language and have minimal awareness of consumer tastes in a non-Chinese setting. More recently, research by Koh (2021) has questioned Forest City’s contribution to local industrial upgrading even further. She discovers the mushrooming of a series of ‘symbiotic’ luxury real estate-related services such as medical tourism, private education, migration advisory, and general high-end consumption. While not intentionally excluding Malaysians, these services seem to be primarily designed to target wealthy Chinese clients. In addition, their scale tends to be modest, which implies an inability to absorb large quantities of local labour. This type of FDI indirectly prevents Iskandar Malaysia from pursuing more sustainable activities with a higher economic multiplier effect such as high-technology manufacturing.

Complicating the situation further is how Forest City was engulfed by the political undercurrents of the host economy. Leading up to Malaysia’s general election of May 2018, Mahathir, then leader of the Pakatan Harapan opposition bloc, publicly attacked the project. Mahathir (2017a, 2017b, 2017c) claimed that the establishment of Forest City has resulted in a substantial transfer of capital and jobs to Chinese companies, along with the influx of Chinese immigrants. He also alleged that the Chinese citizens brought in through Forest City would be accorded voting rights, thereby resha** Malaysia’s ethnocentric political system.

Emphasizing the ‘Chinese-ness’ of Forest City and conflating Chinese investors with the local ethnic Chinese population, Mahathir stirred fears among ethnic Malay voters (Liu and Lim 2019). For some of the more conservative ethnic Malays, who do not always consider fellow citizens of Chinese (and other) heritage as equals, such comments served to undermine UMNO as they relied on the party to counter the influence of the ethnic Chinese minority (see Hamzah 2020; Jaipragas 2017). More fundamentally, projects like Forest City worsen the shortage of affordable housing in Johor. At an average price of MYR1,200 (USD315) per square foot, residential units are effectively priced out of the reach for the majority of Johoreans, especially impacting the ethnic Malay community (Whang 2016).Footnote 11

The above discontent climaxed in the May 2018 general election, whereby UMNO and its allies were rejected by the voters. The party not only lost power at the federal level, but also in ‘safe deposit’ states such as Johor. Analysts attribute UMNO’s loss to various reasons. One of the most commonly discussed factors is the perceived mismanagement of Chinese FDI (see Welsh 2018; Liu and Lim 2019). A survey conducted in February 2018, three months before the general election, revealed that nearly 30% of Johor respondents distrusted and felt disconnected from BRI projects like Forest City. They were uncomfortable with the heavy influx of Chinese investment, which they thought was inflating property prices in the southern state (Chan 2018).

Forest City’s entanglement with domestic politics did not end when Mahathir reassumed the Prime Minister Office (Lim and Ng 2023). In an attempt to close ranks, Country Garden’s founder and then chairman (Yeung Kwok Keung) paid Mahathir a personal courtesy call on 16 August 2018. Yeung also released a press statement shortly after the meeting, proclaiming his confidence in Mahathir’s leadership and intention to deepen Country Garden’s investment in the country (EdgeProp 2018). However, the attempt backfired as Mahathir, in the ensuing days, announced that Malaysia will not allow foreigners to buy residential units in Forest City. He declared that ‘[o]ur objection is because it was built for foreigners… Most Malaysians are unable to buy those flats’ (The Star 2018). When quizzed further, Mahathir clarified that ‘[t]hey can buy the property, but we won’t give them visa to come and live here’ (Today 2018). His administration also proceeded to place Forest City under scrutiny, claiming that a review was due to ensure locals, especially Johoreans, would not be sidelined from buying homes there. Amongst other things, there will likely be stricter limits on foreign buyers at Forest City, but specific details were not revealed to the public (Today 2018).

Discussion

Several points are worth discussing. First, there is a positive correlation between closer Chinese-Malaysian economic integration and the increasing prominence of the services sector in Malaysia. This is especially noticeable from the late 1990s when the services sector expanded almost irreversibly along with deepening China-Malaysia trade (see Figs. 1, 2, 3, 4). Indeed, China became Malaysia’s largest importer and exporter in 2012 and 2019 respectively. To a certain degree, industrial deepening has been exchanged for a large export market in the 2002 signing of the ACFTA, which set the tone for deeper integration with China in the subsequent years. This reliance on China (and other economies) will only escalate following Malaysia’s signing of the Regional Comprehensive Economic Partnership (RCEP) in November 2020 (see Rana et al. 2021).Footnote 12 This finding resonates somewhat with Coxhead (2007), who argues that Southeast Asian economies like Malaysia risk losing comparative advantage in labour-intensive manufacturing to China, in turn trap** these economies in natural resources. While stop** short of agreeing with Coxhead’s depiction of a ‘new resource curse’, the findings of this paper do raise some concerns on Malaysia’s apparent deindustrialization, at least since the late 1990s. Future studies would do well to explore in greater depth the exact mechanism driving such a relationship (if any). For example, a fairly noticeable amount of Chinese FDI has financed manufacturing activities, such as solar panel fabrication, but the chief focus is on low value-added tasks (e.g. assembly and testing) (see Zhang 2021). This potentially stimulates even more debate on the nature of deindustrialization in the Southeast Asian country, going beyond the apparent decline of manufacturing versus the rise of services.

Although it is easy to point the finger at the Chinese economy, there is also a need to scrutinize industrial policies pursued by the Malaysian policymakers. For instance, the mid-2000s initiative to promote economic corridors has not only added another layer of bureaucracy, but also geared increasingly more economic decision-making power to the federal government when it is probably more cost-effective to devolve such powers to the subnational authorities, as Rasiah and Krishnan (2020) and Hutchinson (2015) have argued. The subnational government’s on-the-ground agency arguably provides it with a better vista to make decisions. Equally, the global economy has reconfigured itself considerably since the 1997 Asian financial crisis. In Southeast Asia alone, the growing importance of the CLMV economies since the late 1990s has jacked up the competition to attract FDI, to name one example. Therefore, policies that hitherto worked well for the Malaysians are no guarantee of future success. This finding adds weight to the existing scholarship covering the post-1990s industrialization experience of Malaysia. While crediting the Malaysian state’s relatively open stance towards trade and FDI in the immediate decades post-World War Two, scholars also highlight the diminishing returns of such policies (see, for example, Jomo 2007; Gomez et al. 2021; Rasiah 2011). Additionally, they exhort more carefully calibrated industrial policies to help firms progressively climb the proverbial technological ladder and capture more value from the production and sales of increasingly complex goods and services. For Kam (2013) at least, this means policy focus should extend beyond mere ownership liberalisation and investment attraction. More pertinent are factors that increase productivity such as human capital development and linkage creation. Therefore, a more concerted push in providing, for example, supportive training facilities to facilitate production frontier upgrades rather than the ongoing focus on capital expenditure is worth considering.

Second, from the early 2010s, China–Malaysia economic relationship began to shift to one that is more investment-driven. Following the deceleration of their home economy after the 2008 global financial crisis, increasingly more Chinese firms began to target the Malaysian market. This influx was reinforced in the aftermath of the BRI. While there is some manufacturing FDI from China, the evidence presented earlier on (see Table 2) suggests a heavier presence of tertiary sector FDI. Much of this tertiary sector FDI has financed activities such as real estate, general trade, and broader services. In simpler terms, structural transformation has occurred, but it is cree** towards the services sector. Such ‘easy money’ from China, while not directly jeopardizing Malaysian manufacturing, has almost certainly altered the incentive structure for Malaysian firms, fuelling the Southeast Asian nation’s already noticeable slant towards services, as seen in Figs. 7 and 8. Furthermore, this shift has a lower economic multiplier effect compared to conventional manufacturing activities and has exacerbated social unrest, particularly in the development of Iskandar Malaysia, the nation’s first and most economically consequential corridor. While some argue for the productivity and tradability of these services, the paper does not yet show strong evidence of this. Instead, what has emerged thus far is an enclave-like environment in which speculative activities are promoted.

Building on Koh’s (2021) study, the paper argues that the active marketing of a series of ‘migration products’ not only entrenches Iskandar Malaysia in (unproductive) services (that are difficult to trade across borders), but also fosters interethnic-cum-interclass divisions. This influx of Chinese FDI, while helpful in spurring high-end urban real estate projects, complicates the potential revitalizing of manufacturing by the Iskandar Malaysia authorities. It is true that some service activities (e.g. private education and healthcare) catalysed by projects such as Forest City align with the stated goals of Iskandar Malaysia policymakers. However, this does not substantially change the fact that these activities also encapsulate the regional corridor’s pursuit of ‘soft’ goals over more sustainable ‘hard’ targets. As highlighted by van Grunsven and Hutchinson (2016), the apparent move away from manufacturing (and towards services) in Iskandar Malaysia has already begun even before the massive influx of Chinese FDI. Embracing projects like Forest City further locks-in an ecosystem that does not prioritize long-term, meaningful industrialization, echoing the warnings of Palma (2005) and Rodrik (2016) regarding the risks of deindustrialization in the Global South. This discovery underscores the importance of distinguishing between productive and unproductive (or even exploitative) services. At least from what can be observed from Iskandar Malaysia thus far, Chinese FDI fits the latter category more closely than the first.Footnote 13

The experience of the LAC economies offers valuable comparative insights. If we learnt anything from their relatively low levels of industrialization, it is that such underperformance, to a certain degree, can be explained by these economies’ adoption of Washington Consensus-inspired economic policies. The reluctance to more coherently facilitate value capture from increasingly sophisticated (manufacturing) activities implies that deindustrialization is a virtual certainty. Such policy directives have been in force even before the LAC economies sought closer economic ties with China in recent times, although pursuing the latter has indirectly quickened the pace of deindustrialization. Phrased differently, it is unwise to completely pin one’s deindustrialization woes on warmer economic ties with the Chinese. By the same token, it is still possible for develo** nations to resist or even overturn deindustrialization’s lock-in effect, even if their toolbox is restricted due to issues such as limited fiscal strength.

Conclusion

This paper has forwarded two key arguments. First, it posits that economic cooperation between China and Malaysia may not be the main source of Malaysian industrial deepening. This is evident in the decreased focus on manufacturing and what is traditionally considered ‘proper’ industrialization. Second, Chinese FDI entering Malaysia has largely financed service activities. The services-centric nature of such FDI, especially when concentrated in the host state’s most important economic corridor (Iskandar Malaysia), locks in an ecosystem that already underprovides long-term, meaningful industrialization. Additionally, it sharpens social inequality and political tension.

As if to underline this paper’s core thesis, the situation surrounding Forest City (and other Chinese-financed development in Iskandar Malaysia) has gone from bad to worse. More prosaically, the Malaysian economy, like those of the rest of the world, was soon hit by the Coronavirus outbreak (COVID-19), which caused tremendous suffering. While the nation has reflated its economy somewhat, further progress has been imperilled by events ranging from geoeconomic tensions at the international level to military conflicts in key theatres. In China, the situation is not rosy either, with a looming real estate crisis one of several issues troubling policymakers. At the point of writing, several Chinese real estate companies, led by Evergrande Group (the second largest player in the Chinese market), are facing severe liquidity shortfall (Choong Wilkins 2021). Homebuyer sentiment has also been affected, meaning that sales in their overseas portfolio (e.g. Iskandar Malaysia) will not likely pick up in the near to medium-term. This spells trouble for other emerging economies within Southeast Asia that hitherto have been harnessing real estate dollars from this group of Chinese TNCs.

In view of the events in Iskandar Malaysia and the broader development trajectory of the Malaysian economy, it makes sense for policymakers within the region to rethink their industrial policies. While it is tempting to tap the large Chinese market, especially in the pursuit of services, it is equally important to recognize the potential pitfalls of such a strategy. While Chinese capital exports can catalyse structural transformation, such a process is highly path dependent, which could generate less than satisfactory outcomes. In any case, the growth spurt induced by this form of capital might mask (or even worsen) longer term, structural issues. The key lies instead in revitalizing a nation’s productive structure by executing reforms continuously, which would likely nudge it towards a ‘high road’ of sustainable development, with or without direct participation by Chinese business groups.

Taking a long view of East Asia’s development, is it even prudent to place FDI, from China or elsewhere, at the centre stage of a nation’s industrialization ambitions? A careful analysis of East Asia’s post-World War Two development reveals that virtually all the high performers, except city-states like Hong Kong and Singapore, did not rely on FDI to stimulate their productive capabilities. If anything, it can be argued that Malaysia’s middle-income trap has, to some extent, been caused by an overdependence on its FDI-friendly strategy. While foreign money and expertise helped Malaysia attain a relatively high growth rate, the same approach has lost its lustre by the late 1990s, a point raised earlier on but worth reiterating here. Tangential but important to this discussion is the extent by which these East Asian high performers devote policy attention towards grooming manufacturing firms. Several of them have since established forte in their respective fields, ranging from automobile (Japan’s Toyota), consumer electronics (Korea’s Samsung), to bicycle (Taiwan’s Giant). Even Singapore, despite its generally laissez-faire economic approach and sophisticated service industries, consistently seeks to attract and retain manufacturing activities deemed critical to national interests. Common examples include semiconductors and advanced medical products. Hong Kong is the exception to the other economies mentioned here, but it is worth recalling that its economic take off in the immediate years post-World War Two was partly sustained by manufacturing expansion.Footnote 14

The overriding lesson here is that a nation’s true source of sustainable development is continuous value creation by domestic firms, especially in progressively technology-intensive manufacturing, as documented by Cherif and Hasanof (2019) and Ohno (2009). This is not to suggest that Malaysian (and by extension, Southeast Asian) policymakers stop attracting FDI altogether as that would be unrealistic. However, there is merit in fostering a stronger connection between domestic innovation and the long-term upgrading imperatives of TNCs (see, for example, Kam 2013). Put another way, the goal is not simply to bring as many investors as possible into the country, but to continuously embed them into the domestic organizational ecology in a mutually beneficially ‘upgrading coalition’ (Doner and Schneider 2016). Parallel to this is the need to more systematically target FDI with dense positive externalities, while also phasing out unproductive, speculative ones that do not directly contribute to domestic capability building.