Abstract
Since the 1970s development studies have conveyed an impression of Taiwanese firms as being active small and medium-sized enterprises (smes) with flexibility to successfully survive in a competitive global market. On the contrary, we use the unbalanced panel data of 2,969 top manufacturers during 2002–2015 to explore why and how Taiwanese firms expanded their scale and scope of operations in the new century. Our findings indicate that Taiwanese subcontractors are caught in a dilemma between the expansion of operational scale in China and industrial upgrading in Taiwan. Scale expansion can exploit the large cheap labour pool in China with minimal effort for most smes but significantly reduces the firms’ profit rates. Industrial upgrading is capital-intensive and profitable for a few Taiwanese firms but challenging for most smes to attract long-term investment in research and development. The ‘subcontractors’ dilemma’ explains Taiwanese firms’ struggles for survive in the US–China trade conflicts.
1 Introduction
On 27 July 2017 at the White House, Donald Trump, the president of the United States, together with Terry Guo, the chairman of the Hon Hai Precision Industry Taiwanese business group (operating under the name Foxconn), announced a plan to invest US$10 billion in a new factory expected to employ 3,000 people in the state of Wisconsin. 1 President Trump claimed credit for Foxconn’s ‘incredible investment’ and boasted of reviving the US manufacturing sector and ‘making America great again’ ( bbc News, 2017). Foxconn, the major producer of iPhones and iPads for Apple and notorious for operating sweatshops in China, is the world’s third largest employer (with approximately 1 million employees in 2017), following Walmart (2.2 million) and China National Petroleum Corporation (1.7 million) (Fortune, 2015). The dramatic announcement of Foxconn’s investment in the United States reflected the considerable scale and influence of major Taiwanese companies in the global electronics supply chain. One year before announcing it was establishing operations in the United States, Foxconn frequently made headlines in Japanese newspapers following its acquisition of Sharp Corporation (sharp), which, established in 1912, is the 10th largest television manufacturer in the world, the third-highest-volume producer of mobile phones in Japan, and employs 50,000 individuals. The Foxconn–sharp acquisition deal surprised the Japanese public because of a stereotype that had emerged from development studies in the 1980s (e.g. Hamilton, 1997): that Taiwanese firms tend to only be small and medium-sized enterprises (smes). This outdated impression of some Japanese reflects an underestimation of the current scale and influence of Taiwanese firms.
The influence of Taiwanese multinational corporations such as Foxconn has profoundly challenged this stereotype. In the new millennium, some of the most dramatic changes undertaken by Taiwanese manufacturers have been expanding workforces and increasing the scale of revenue. News concerning major Taiwanese companies, such as Foxconn and the Taiwan Semiconductor Manufacturing Company (tsmc), has frequently made headlines in the global business media. However, erstwhile studies of firm expansions and survival strategies were biased towards the most successful cases, such as tsmc, which was significantly supported by the industrial policies of the state (Chu & Amsden, 2003). The story of how Taiwanese smes such as Foxconn became some of the greatest ‘world factories’ has largely been neglected. 2 Therefore, explanations for why Taiwanese companies increased their scale and scope still elude us.
During the past two decades, Taiwanese smes have undergone a marked growth in scale. As Figure 1 illustrates, in 2013 the average number of employees, including overseas employees, at the top 10 Taiwanese business groups was approximately 196,000 ( CommonWealth Magazine, 2016), which was greater than that of Honda (approximately 179,000 workers), the second-largest automobile manufacturer in Japan (Fortune, 2015). Most of the overseas employees at the largest Taiwanese companies are Chinese migrant workers who work on labour-intensive assembly lines in the southern and eastern coastal provinces of China (Lin et al., 2016). Taiwanese smes transformed into industrial giants over a span of approximately two decades because of the cheap labour and land provided by China, and they are now striving to upgrade their technology by merging with upstream Japanese companies such as sharp.
Employment in the top 10, 100, and 1,000 enterprises in Taiwan.
Citation: International Journal of Taiwan Studies 2, 2 (2019) ; 10.1163/24688800-00202002
Source: CommonWealth Magazine (2016) database of the top 2,000 Taiwanese enterprises.Scholars of economic development theory have applied the concept of ‘flexible production’ embedded in the social network to explain the prosperity of Taiwanese smes since the 1980s (Hamilton & Biggart, 1988; Hamilton & Kao, 1987). The argument goes that smes could adopt global market competition effectively because of their flexibility in networking and learning-by-doing with specific technical niches rather than scaling and internalising the technical core as large companies do (Friedman, 1988; Granovetter, 1985; Piore & Sabel, 1984). Nevertheless, the transformation of the scale and scope of Taiwanese firms has created both theoretical puzzles and empirical anomalies for the flexible production theory (Lee & Hsiao, 2014; Lin, 2015).
In this paper, we challenge the flexible production perspective, which failed to predict the extraordinary transformation of Taiwanese firms during the past two decades, and attempt to apply a new perspective to explain this unprecedented expansion in scale and scope. We used unbalanced panel data for 2,969 manufacturers from the database of the top 5,000 Taiwanese firms during 2002–2015 to corroborate the theory that the expansion of Taiwanese firms is a strategy for subcontractors of supercapitalism (soss, abbreviated as such because they usually struggle to survive in the global market). This strategy relies on a rapidly increasing workforce in China, which generates a low profit rate (Lin & Hu, 2017). These subcontractors rarely adopt the specification strategy, which relies on heavy investment, expensive skilled workers, and industrial upgrading; exceptions include tsmc and a few smes that have kept operations in Taiwan (Chu & Amsden, 2003; Hsieh, 2014). The statistical results indicate that subcontractors face a dilemma between engaging in short-term scale expansion in China and undertaking long-term industrial upgrades in Taiwan.
The subcontractors’ dilemma has not only changed the industrial structure in Taiwan but also resulted in some unintended political consequences. The strategy of horizontal integration in China not only boosted the scale of Taiwanese companies such as Foxconn but also shaped those companies’ attitudes and power in the cross-strait relationship. Taiwanese business groups that rely heavily on investment and employment in China prefer a more open cross-strait trade policy than local smes and their employees. Requests by Taiwanese business groups for government favours on cross-strait trade and tax policies 3 have triggered political discontent and a spirit of economic nationalism among Taiwanese citizens who worry about China’s political influence, culminating in support for the Sunflower Movement, a large-scale protest that challenged the cross-strait open trade policy (Ho & Lin, 2015). Although the movement triggered changes in the trend of political interactions between Taiwan and China, it has not changed the Taiwanese subcontractors’ dilemma, which remains embedded in the global supply chain and cross-strait economic interdependence that results from the competition of supercapitalism. We will discuss the implications for industrial and trade policies of our findings in the conclusion.
2 Theories of Firm Size and the Taiwanese Anomaly
Firm size is a long-established concern in economics and sociology. The conventional firm size theory argues that following advancements in technology and market saturation, the relatively small size of smes may restrict their capability to market competition (Chandler, 1977). Firm size is relevant because of three mechanisms that determine the profit margins of firms. First, industries with a scale economy, which refers to the threshold of substantial sunk investment (capital-intensive) and low marginal unit cost, may exclude smaller-scale newcomers from the market. Therefore, scale economy industries, such as those of electric power, telecommunications, transportation, and water systems, typically determine the horizontal integration and ‘natural monopoly’ structure of the market (Sharkey, 1982). Second, for industries with high quality-control costs and information asymmetry, the transaction cost between upstream and downstream companies through the market may be higher than an organisation’s cost of hierarchical integration and direct management (Coase, 1992; Williamson, 1981). Third, the conventional wisdom of organisation theories assumes that, on average, firm size—scale and scope—increases, and large companies with efficient bureaucracies would be capable to drive small companies out of the most mature markets in the long run (March & Simon, 1958). In his historical study, Chandler (1990) argues that three economic mechanisms, scale (horizontal integration), scope (vertical integration), and geographic expansion (internationalisation), are three major strategies for modern corporations to gain advantages over their competitors. In addition, in contrast to vertical integration (scope), he argued that in the case of adopting horizontal integration (scale) as a strategy ‘resulting cost advantages were minimal’ (1990: 38).
Neo-Marxists agree with the historical trend of monopoly capitalism but from a different perspective. They have argued that because of the cyclical crisis of overproduction and declining profits, capitalist states should use monopoly capitalism to stimulate mass consumption and hedge against capitalism crises (Baran & Sweezy, 1966). Using Keynesian fiscal tools, welfare policies (O’Conner, 1973), and monopoly capitalism, state intervention in capitalist economies has reorganised industries according to the Fordist mode of production, which refers to a Chandlerian bureaucratic organisation with high wage levels and adequate welfare for the employees attained through collective bargaining with trade unions (Burawoy, 2005/1979). For the neo-Marxists, Fordism successfully enabled the avoidance of the chaos of competitive capitalism and initiated a golden age for welfare states and mass production in Western countries after World War ii (Offe, 1984).
Fordist or Chandlerian firms, however, have been theoretically and empirically challenged since the late 1970s. Empirically, neoliberal economic policies in the West have undermined welfare states since the 1980s (Block & Evans, 2005). Moreover, monopoly capitalism transferred production from the Rust Belt region in the northeast United States and European countries to other emerging economies worldwide (Babb, 2005). Theoretically, some scholars have observed the prosperity of smes in certain industrial areas, such as northern Italy and Taiwan, highlighting the advantage of flexible production systems and social networks (Piore & Sabel, 1984). In contrast to their views on the Japanese zaibatsu and the Korean chaebol, the advocates of the flexible production theory contend that Taiwanese manufacturers exemplify the ability of smes to successfully adapt to global market competition (Hamilton & Kao, 1987; Hsieh, 2011). According to the theories aforementioned, smes have the advantage of active entrepreneurship, which, when embedded in social networking, can facilitate the distribution of skills and information and the reciprocity of informal finance that may help smes compete with large businesses (Castells, 2000; Gereffi & Wyman, 1990; Hsieh, 2014).
However, these theories of scale and scope cannot sufficiently explain the extraordinary transformation of Taiwanese firm size over the past two decades. First, the largest companies in Taiwan are mostly active in the electronics and footwear industries, which have traditionally been labour-intensive rather than capital-intensive. Thus, the scale economy may not be applicable to these industrial giants. Second, with the progress of Internet technology and the advent of globalisation, the transaction cost between enterprises may not have increased but rather decreased (Castells, 2000; Lamoreaux, Raff & Temin, 2003). Some multinational enterprises, such as Apple and Nike, have adopted vertical disintegration strategies such as subcontracting in order to reduce organisational costs and manage the uncertainty of global market volatility (Gereffi, 1994; Gereffi, Humphrey & Sturgeon, 2005; Gilson, Sabel & Scott, 2009; Langlois, 2003). However, this does not explain why their Taiwanese subcontractors, like Foxconn, returned to horizontal and vertical integration strategies in the age of the Internet. In the era of neoliberalism, the neo-Marxist explanation of the transition from competitive capitalism to monopoly capitalism has been reversed in practice (Burawoy, 1985) and is irrelevant in explaining the increasing size of Taiwanese firms in China. As some studies of world factories have noted, these companies usually exploit cheap labour and expropriate farming lands from poor peasants without providing sufficient compensation (Lin et al., 2016). Moreover, the flexible production perspective cannot clearly explain why some Taiwanese smes elect to renounce their advantage of being embedded in social networks (Shieh, 1992; Uzzi, 1999) and establish mass production in China. Thus, we cannot depend on conventional explanations and must instead develop an alternative theory to explain the evolving institutional environment, the reaction of firms to market conditions, and the expanding size of Taiwanese companies.
3 Subcontractors’ Dilemma
When Chandlerian-style enterprises saw their market ranking begin to deteriorate and their reputation criticised as the ‘vanishing hands’ in developed economies (Lamoreaux, Raff & Temin, 2003), why then did Taiwanese smes seemingly seek to follow their model by giving up their advantage of flexibility and expanding in scale and scope? The answer, we argue, is that the increased size of Taiwanese companies is a consequence of subcontractors’ weakened leverage in the global value chain because of the intense competition of supercapitalism. In his book Supercapitalism, Reich (2007) reveals that the average economic scale of the top US companies continually increased in the era of globalisation, but their survival rate decreased. According to Reich, supercapitalism generates relatively intense competition because of short-term profit pressure from stockholders in the financial market and shorter life cycles of products in the consumption market, thereby resulting in the ‘creative destruction’ of firms on a greater scale than before (Schumpeter, 1994/1942). In the age of supercapitalism, a few leading companies in innovative economies receive large profits as the reward for risk-taking and capitalising on opportunities from financial and consumption markets, while others struggle for survival. Most Taiwanese firms fall within the second category.
Taiwan’s CommonWealth Magazine (2018) ranked the top 1,000 companies in the Greater China region (i.e. China, Taiwan, and Hong Kong) according to revenue in 2018; the ranking of Taiwanese companies followed the same pattern as that of the US companies reported by Reich (2007). Although the average size (revenue) of Taiwanese companies increased in an extremely short period (early 2000 to 2015), the rankings of most Taiwanese firms declined. Moreover, the total number of Taiwanese firms in the top 1,000 list continually decreased from 2006 (355 firms) to 2018 (124 firms). This decline highlights the disadvantages of Taiwanese firms operating under intense market competition in the Greater China region.
Supercapitalism theory may explain the increasing pressure of global and regional markets but it cannot sufficiently explain why firms in the same supply chain adopt different management strategies to change their sizes. Therefore, we contend that the level of bargaining power of firms in the global value chain determines their size (Gereffi, Humphrey & Sturgeon, 2005). Brand-name firms that control marketing, innovation, and core technology acquire maximum profits by using the Internet and the international subcontracting system to reduce manufacturing costs (Lin et al., 2016). Thus these brand-name firms become the vanishing hands in that they tend to expand their research and marketing and downsize their manufacturing in developed countries (Langlois, 2004). For example, Apple hired 97,200 employees (mostly in the United States) in 2015, which amounted to less than 10 percent of the number of employees working at their subcontractor Foxconn in that year (Fortune, 2015). However, if brand-name firms identify the competitors of their subcontractors, they can strengthen their bargaining power in the supply chain, which compels the subcontractors to participate in a race to the bottom under market competition.
Table 1 presents a comparison of the basic management data of Samsung, Hon Hai (Foxconn), and tsmc, the three most prominent firms in the East Asian electronics industry. The experiences of these three companies reveal that subcontractors adopted three major strategies (vertical integration, horizontal integration, and industrial upgrading) to survive or improve their leverage in the global value chain.
The most effective strategy to prevent weak bargaining power and thin profit margins is to compete with global brand-name firms; small firms that execute this strategy successfully may themselves become global brand-name firms. The Samsung group, a typical Korean chaebol, was once, between 1970 and 1980, a subcontractor of Intel. Because of its large size, monopoly on the domestic market, and complex relationship with the government, Samsung adopted the effective strategy of investing in marketing, capital, and research (Samsung Newsroom, 2019). For example, it increased its workforce to approximately 320,671, nearly equalling Toyota’s employment (344,109), and generated similar revenues to the Japanese automaker. The chaebol-isation strategy requires long-term financial support and research funding from both banks and the state (Amsden, 1989). This vertical integration strategy has effectively improved the performance of Korean firms; for example, in 2017, Samsung attained a substantial profit rate (17.3%) that was notably higher than that of Foxconn (4.5%). However, Taiwanese smes face difficulties adopting such a strategy because it requires considerable investment in research and marketing paid for by state funding and loans from banks. Although Taiwan under the Kuomintang (kmt) government prior to democratisation was considered an ideal case of a developmental state (Deyo, 1987; Haggard, 1990; Johnson, 1982; Wade, 2004/1990), the kmt government questioned the political loyalty of local smes and limited the extension of loans from the state-owned banking systems to them before 1990 (N. Wu, 1987; Y. Wu, 2004). Therefore, only a few Taiwanese smes, such as the bicycle manufacturer Giant, could seize the market opportunity, adopt the strategy of increasing scope, and successfully shift from subcontractor to global brand-name firm (Hsieh, 2015).
As some studies of flexible production have noted (Hsieh, 2014), an alternative strategy through which a subcontractor might increase its bargaining power and profit rate is the industrial upgrading strategy, which entails specialising a core technology in the supply chain. tsmc, which has been supported by kmt governments and initially received public funding for research, may be the most illustrative case that demonstrates the benefits of adopting this approach in the long term (Chu & Amsden, 2003; Mathews & Cho, 2000). The scale of tsmc’s revenue and employment (48,602 employees in 2017) is much smaller than that of Samsung and Foxconn, and most tsmc employees are Taiwanese engineers. However, tsmc has the highest stock market value in Taiwan and depends heavily on cash flow from its foreign investments through the financial market. Because of the high capital expenditure per employee ($223,500 in 2017) and having the highest market share in the global semiconductor foundry market (54% global share in 2015), tsmc has earned a considerable profit margin, which has been on average approximately 30 percent, much higher than that of Samsung and Foxconn during the past decade ( CommonWealth Magazine, 2016). Nevertheless, because of the considerable investment required and long-term uncertainty, only a few Taiwanese subcontractors have adopted an upgrading strategy.
The final and most feasible strategy for soss is the rapid expansion of scale. As exemplified by Foxconn, subcontractors may pursue horizontal integration by absorbing production factors, such as cheap labour and land in China (which offers the largest scale of production worldwide), in order to reduce production costs and the number of potential competitors, particularly in instances when establishing a brand name and pursuing industrial upgrading would be expensive and high-risk. Although subcontractors like Foxconn can adopt vertical integration with the consent of valuable customers such as Apple, they may not be able to challenge brand-name firms in the major markets without incurring punishment from their customer companies (Lin et al., 2016). Therefore, although increasing corporate scale and racing to the bottom of prices and production costs might reduce the net profit rate, it is a feasible survival strategy for low-end subcontractors in the short term.
The subcontractors’ dilemma refers to the trade-off between, on the one hand, strategies that pursue scope or upgrading, which mean taking risks on marketing or innovation and generating higher profit rates in the long run, and on the other hand, strategies that pursue scale for the sake of low risk and low profit margins without upgrading in the short run. Although the Chandlerian model (1977, 1990) has been applied by Chu and Amsden (2003) to explain the development of the Taiwanese electronics industry, it does not account for this dilemma, which is derived from the weak power leverage of subcontractors in a late developed country in a global supply chain shaped by supercapitalism.
To apply our theory of the subcontractors’ dilemma, we initially divided our sample of 2,969 Taiwanese companies operating between 2002 and 2015 into three groups: first, continuous smes [1]; second, profitable minor large firms that adopted vertical integration (mostly comprising continuous large firms) [2]; third, firms that adopted horizontal integration by employing cheap migrant labour in China. As a consequence of their strategy, firms in the third group faced two potential outcomes. First, they changed from smes to large business groups [3] due to their expansion in China. Second, they became large companies in China at the expense of reducing their size in Taiwan [4]. Thus we were left with four groups of firms, a group of large firms and three groups of smes, to demonstrate the disadvantages faced by some smes in Taiwan:
- [1] continuous smes (57% of the sample)
- [2] continuous large firms (24.9%)
- [3] sme-to-large-firm (9.5%)
- [4] large-firm-to-sme (8.7%)
Hypothesis 1 (H1): Taiwanese smes’ profit margins suffered more than the large companies’ profit margins because of competition arising from supercapitalism.
Hypothesis 2 (H2): The greater dependence on Chinese labour of Taiwanese smes, especially large-firm-to-sme, reduced their scale in Taiwan and increased their scale in China, resulting in lower profit margins.
Levels of employment in China of the four groups of Taiwanese manufacturing firms (2002–2015).
Citation: International Journal of Taiwan Studies 2, 2 (2019) ; 10.1163/24688800-00202002
Net profit rates of the four groups of Taiwanese manufacturing firms (2002–2015).
Citation: International Journal of Taiwan Studies 2, 2 (2019) ; 10.1163/24688800-00202002
Hypothesis 3 (H3): Vertical integration measured by consortium assets and employees in Taiwan increases the profit margin of a firm.
Hypothesis 4 (H4): The higher investment on industrial upgrading of the Taiwanese firms measured by assets per employee is, the higher the profit margin they earn.
Therefore, we contend that when smes face the challenge of supercapitalism (H1), the strategy of expanding employment in China (H2) is the easiest and most popular approach adopted by Taiwanese companies looking to ensure short-term survival at the expense of a higher profit margin. By contrast, we argue that the second and third strategies, vertical integration and industrial upgrading (H3 and H4), are less commonly adopted but are more profitable for Taiwanese firms in the long run.
4 Data and Method
Data sources: We empirically tested the four hypotheses using Taiwanese firm-level data. The unbalanced panel data of the top 5,000 Taiwanese companies from 1995 to 2017 were originally collected by the China Credit Information Service (ccis) in 2018, which provides information regarding companies to banks and for academic research in Taiwan. We verified the ccis data by matching them with data from CommonWealth Magazine (2016). We rectified certain errors and retained 2,969 companies from the manufacturing sector; some companies belonged to the top 550 largest business groups in Taiwan during 2002–2015. Indeed, the ccis collected some information on these firms from 1995 to 2017. However, we realised that certain data regarding the consortiums’ assets and workforce in China before 2001 were missing. Officially, Taiwanese companies were not legally allowed to invest in China until 2002 because of national security concerns; however, considerable amounts of undocumented and unapproved investments occurred in China that passed through third parties. Moreover, some companies’ information was not completely released during 2016–2017. Therefore, for data reliability, we focused on the data set from 2002 to 2015 and calculated the means of workforce in China for each of the four groups. The data set included the identity, revenue, age, profit rate, assets, and workforce of the 2,969 companies, the assets and workforce of their parent business groups (consortiums), and the assets and workforces utilised for their consortiums’ investments in China. The company–year panel was unbalanced and included 21,071 observations with complete information on every variable we applied in the models.
Dependent variables: We used two variables to measure the performance of each company. We applied the net sales (revenues) to measure size and applied the net profit rates (after tax) to measure the profit margin. The two dependent variables were used to test the four hypotheses.
smes (flexibility): To test H1, which challenges the flexible production perspective, we assumed that most of the smes that failed to improve either revenue or profit rate decided to relocate to China and transform into large enterprises. We defined smes as companies with fewer than 300 employees on their annual record (smes = 1). The reference group here was large companies (smes = 0). Additionally, we applied four dummy variables—continuous smes, sme-to-large-firm, continuous large firms, and large-firm-to-sme—to estimate whether the changing scale of a company influenced its profit margin.
Horizontal integration (scale): We demonstrated that relocation to China and horizontal integration were the major strategies of the smes that increased their average revenue. We assumed that the greater a firm’s consortium’s investment in assets and workforce in China, the higher the firm’s revenue and the lower its profit rate. In addition, we proposed that the large-firms-to-smes group that mostly reduced their scale in Taiwan and expanded their consortiums’ workforce in China would attain the lowest profit rate of the four groups (H2).
Vertical integration (scope): As stated in H3, we assumed that the higher the degree of vertical integration in Taiwan, the higher the profit rate of the company. We used the firms’ investment in assets and workforce in a specific consortium to estimate the advantage of vertical integration. Therefore, in Taiwan, belonging to a consortium that is considered large in terms of assets and workforce improves the profit rate of the company.
Industrial upgrading: We used each company’s assets per employee in Taiwan to measure industrial upgrading (H4). As illustrated by the tsmc case, we assumed that higher assets per employee for a company in Taiwan means a higher probability that the company performs adequately in its specific product market and therefore that it increases its profit rate after controlling its revenues.
Control variables: Conventional economic models have assumed that increasing the assets and number of employees of a company leads to a reduction in the marginal production possibility of their revenue. To demonstrate the nonlinear relationship of assets and employees on revenue and profit rate in our models, we introduced their squares in both individual firm-level and consortium-level variables. We also introduced two control variables: the company being listed on the stock market (yes = 1) and the age of the company. With reference to some studies of firms’ scales and their mortality (Daepp et al., 2015), we applied the square of age in our models. The descriptive statistics of the selected variables are presented in Table 2. In Table 3, we present a simple correlation matrix of the variables.
Figure 4 and 5 are line charts of the changing average assets in Taiwan, consortium assets in Taiwan, consortium assets in China, workforce in Taiwan, consortium workforce in Taiwan, and consortium workforce in China. Figure 4 shows the effects of the six variables on revenue (net sales); Figure 5 shows their effects on net profit rates. Figure 4 shows a general linear relationship with slightly negative marginal effects on a firm’s revenue from consortium assets, assets in China, and workforce in Taiwan; and with slightly positive marginal effects from the company’s assets, consortium workforce, and workforce in China. In brief, the greater the scale of capital and labour of a company and its consortium in both Taiwan and China, the greater its revenue, except for the slightly marginal effects resulting from each variable.
Effects of assets, consortium assets, assets in China, employees, consortium’s employees, and employees in China on the net sales of the four groups of Taiwanese manufacturing firms (2002–2015).
Citation: International Journal of Taiwan Studies 2, 2 (2019) ; 10.1163/24688800-00202002
Effects of assets, consortium assets, assets in China, employees, consortium’s employees, and employees in China on the net profit rates of the four groups of Taiwanese manufacturing firms (2002–2015).
Citation: International Journal of Taiwan Studies 2, 2 (2019) ; 10.1163/24688800-00202002
Were these scaling strategies profitable for the companies? In Figure 5, we present the complex relationships of the six variables on net profit rates. First, the companies’ increasing assets, workforce, consortium workforce, and workforce in China result in a marginal decline to their profit rates. Only the consortium’s assets, which were applied to measure vertical integration, exhibited continuously improving profit rates. The reach of the inverted-U shape of the consortium’s assets in China and workforce in China were especially positive skewed and close to zero, which implies that the expansion of capital and labour in China usually reduces the net profit rates and continuously races to zero.
Revenue(Profit) it = Revenue mean of i(t − 3) ~ (t − 1) + X it β + t dummy + γ i + ε it
for t = 1, 2,…14 and i = 1, 2,…2,969.
where the dependent variable was observed for company i at time t, X it is the time-variant 1 × k regressor matrix, γ i is the unobserved random effect of the firm i, and ε it is the error term. The year dummy variable t was used to control for the general annual trend of the economic cycle.
5 Analysis
As shown in Table 4, we used the random effects regression models with the year dummy from 2002 to 2015 (2002 = reference) to analyse two dependent variables: net sales in Models (1) and (2) and net profit rates in Models (3) and (4). We applied the single dummy variable smes in Models (1) and (3), and the other three dummy variables—sme-to-large-firm, continuous large firms and large-firm-to-sme—in Models (2) and (4), taking the continuous smes as the reference group. The overall R-squared of the two models for estimating revenue was as high as 0.964. By contrast, the overall R-squared of final Model (4) for estimating net profit rates, as presented in Table 4, was only 0.02. The low R-squared value of the net profit rates in Models (3) and (4) may indicate that numerous strategies adopted by Taiwanese companies and consortiums did not resolve the problem of their low profit rates but were useful for maintaining the companies’ annual revenues and persevering under global pressure.
The results in Model (1) indicated that the smes had negative effects on net sales during 2002–2015. This result confirmed our argument that Taiwanese smes’ revenue declined because of global market competition (H1). We then divided the dynamics of companies into the other three groups with reference to the continuous smes in Model (2). The results revealed that the large companies continuously increased their revenue over the 14 years. Results of applying the control variables in Models (1) and (2) demonstrated that both the increase of the firm’s and its consortium’s assets led to a positive direct effect and a negative marginal effect on its revenue. By contrast, an increase of a firm’s and its consortium’s workforce led to a negative direct effect and a positive marginal effect on its revenue. In addition, an increase of consortium assets in China led to the reduction of revenue, but the increase of consortium workforce in China led to the expansion of revenue. The results matched the effects of another key variable—assets per employee—that was negatively associated with revenue. Although listed companies are usually bigger, the age of a company led to an inverted-U relationship to its revenue. All the results implied that the booming revenue of Taiwanese companies mostly came from increasing the workforce, especially in China.
The results of Model (3) indicated that being an sme had significant negative effects that reduced the net profit rate by 4.7% (H1). According to the results of Model (4), the large-firms-to-sme group, which on average employed more workers in China, led to a major reduction of 14.7% in the net profit rate of the four groups. The control variable of revenue in the last three years was another indicator of horizontal integration that also significantly reduced the profit rate. Additionally, the coefficient of employees in China exhibited an inverted-U shape with regards to its influence on profit rate, suggesting that the companies’ profit rate declined when their consortium hired more than nine workers in China. In short, the expanding revenues of companies were mostly because of increasing employment in China, especially for companies that reduced their workforce in Taiwan and shifted from operating as large enterprises to smes, which damaged their profitability during 2002–2015 (H2).
To cope with the competition entailed by supercapitalism, some companies adopted a vertically integrated strategy or an upgrading strategy. The results of Models (3) and (4) indicated that the greater the consortium assets of a firm, the better its profit rate, confirming that the strategy of vertical integration enhanced net profit rates (H3). Second, some companies maintained their competitiveness by adopting the upgrading strategy (H4). The upgrading approach focuses on asset expansion and entails the acquisition of a high concentration of fixed assets or physical property, particularly for advanced processes mainly used to produce semiconductor-related products; these processes are employed by tsmc. The results of Models (3) and (4) also confirmed that the higher assets per employee measured in industrial upgrading had a positive effect on the net profit rates after their net sales and other variables had been controlled.
6 Conclusion and Discussion
In this paper, we described the transformation of Taiwanese firms over the past two decades. We demonstrated that the scale and scope of Taiwanese companies are expanding and that some companies are currently among the largest globally in terms of revenue and employment. The large sizes of these top firms invalidate the stereotype that Taiwanese companies can only be smes. Two decades ago, the majority of these industrial giants enlarged their companies by investing and expanding their workforces in China. We need a new theory to explain why and how so many Taiwanese manufacturing companies adopted different strategies to enable major expansions.
The enlargement of Taiwanese firms disproves the theoretical explanation of firm size. The general theory of firm size focuses on industries with the characteristic of economies of scale; however, the largest Taiwanese firms do not belong to such industries. The transaction cost theory explains global brand names’ downsizing strategies rather than subcontractors’ upsizing strategies. Regional studies of Taiwan’s economic ascent have typically claimed that the flexible network of smes might confer an advantage for Taiwanese companies that compete with large companies in the global market. Indeed, the flexible production perspective might still effectively explain smes’ outstanding performance in some manufacturing sectors such as the machine tool industry (Hsieh, 2014). However, it is difficult to explain how so many smes transformed into industrial titans in such a short period of time.
We suggest that the power imbalance between global brand-name firms and subcontractors in the highly competitive age of supercapitalism has shaped the strategic dilemma of Taiwanese smes operating as the latter. These smes applied three survival strategies: horizontal integration (scale), vertical integration (scope), and upgrading (specialisation) in the global market competition. Before the 1980s, private entrepreneurs in Taiwan could not consider vertical integration without political and financial support from the kmt party-state and the state-owned banks. However, after the 1980s, the government selected only a few ‘high-tech’ enterprises, such as tsmc, to which to provide public research funding and financial support. Therefore, most smes retained the subcontracting orders from their brand-name customers by maintaining low prices and reducing costs through investing in China and expanding the scale of employment sharply in the past two decades.
We used the top 5,000 Taiwanese companies’ panel data for 2002–2015, revised from the database collected by the ccis, to test our hypotheses concerning the subcontractors’ dilemma. The results indicated that of the 2,969 manufacturing companies, the smes comprised three groups, some continuous smes, others that changed from smes to large business groups because of expansion in China, and others that became large companies in China at the expense of reducing their size in Taiwan. The last group earned significantly lower profits than those of the continuous smes on average. The trends of firm size in the various groups confirmed our theoretical assumptions, and the statistical results of the random effects regression models of the unbalanced panel supported our hypotheses.
Why did most Taiwanese firms expand sharply but still fail in global competition? Our findings indicated that firms continuously investing in Taiwan and pursuing vertical integration or upgrading strategies without dramatically expanding their scale in China usually achieved higher profit margins over the long run. By contrast, firms that invested heavily in China might have survived in the short run but reduced their profit rates over the long run. The reality of these low profit rates earned by these scaled-up companies challenges the argument that Taiwanese firms should increase their scale in Taiwan and China in order to promote upgrades in the global supply chain (Chu & Amsden, 2003). Scaling might instead be a substitute for upgrading.
Although the strategy of increasing scale in China might not be the optimal choice, it is a popular approach for soss. This popular strategy has increased Taiwan’s economic dependency on China, thereby reshaping political cleavages in Taiwan over the past two decades (Lin, 2015). Because of their considerable investment in China, some large Taiwanese companies, such as Foxconn, have become supporters of kmt politicians and their pro-China economic policies. After Ma Ying-jeou won the presidential election in 2008, the kmt government implemented a series of bold liberalisation policies concerning Taiwan and China; for example, in 2008, the government permitted regular direct flights between the countries, and in 2011, mainland tourist visits to Taiwan were permitted. The highlight of President Ma Ying-jeou’s efforts to improve cross-strait relations was the 2010 bilateral Economic Comprehensive Framework Agreement (ecfa), which sought to minimise barriers to cross-strait trade and envisioned a scheduled roadmap for additional negotiations concerning investment protection, merchandise trade, and service trade. Three years later, in June 2013, the Cross-Strait Service Trade Agreement (cssta) was signed and political furore ensued (Ho & Lin, 2015).
Ironically, the Taiwanese capital flight to China declined after the ecfa began. For the former smes, horizontal integration by expanding workforce in China might have been a popular survival strategy, but it could not save these Taiwanese firms from low profit rates under supercapitalism. Since 2010, as Figure 6 shows, the percentage of Taiwanese foreign direct investment flight to China declined from 83% to 44%, which was close to the record low set in 2001. The ‘salmon returning home’ (guiyu fanxiang) policy for attracting returnee investment from China-bound Taiwanese business groups was first devised in 2006, but yielded significant results only after 2012, partly because the kmt government offered greater incentives and partly because of the rapid increase in production costs and labour unrest in the coastal provinces of China (National Development Council, 2016). 4 The increase in returnee capital was appreciated because it was expected to create new job opportunities and populate Taiwan’s stagnant industrial zones. Nevertheless, the returnee business groups maintained their operations in China, which helped Beijing to gain leverage over Taiwan’s domestic politics (Ho & Lin, 2015). 5 The management style of returnee business groups, which notoriously involves sweatshops in China (Lin et al., 2016), has aggravated labour concerns regarding these recent returnee capitalists in Taiwan.
Taiwanese foreign direct investment flight to China, other areas of Asia, Japan, and the US (1991–2017).
Citation: International Journal of Taiwan Studies 2, 2 (2019) ; 10.1163/24688800-00202002
Note: Taiwanese foreign direct investment flight fell to 44% in 2017, close to the level of 2001, from a peak of 83.8% in 2010.Similar to the discontent towards globalisation that is evident in other industrialised countries (Rodrik, 2011; Piketty, 2014), the transformation of Taiwanese business groups and the cross-strait open trade policies incited political discontent against large companies. In March 2014, the Sunflower Movement, a student-led protest to oppose the cssta, shocked the public. For 24 days (18 March–10 April 2014), Taiwan’s legislature was paralysed by protesters occupying the conference chamber, while others camped in the surrounding roads and alleys to prevent police eviction. The protesters claimed that the cssta would favour large companies investing in China and that the cssta would damage local smes and have a deflating effect on local labour’s wages. Some famous local entrepreneurs, such as Rex How, openly echoed these claims. 6 Observers believed that the Sunflower Movement empowered the Democratic Progressive Party (dpp), which won the national election in 2016 (Ho & Lin, 2015). However, the recent failure of the dpp administration in the local elections of 2018 may indicate a strong realignment of pan-kmt interest groups, which include Taiwanese businesses that still deeply depend on China. To protect the huge businesses in China, Terry Guo, the chairman of Foxconn, announced in April 2019 that he would run for Taiwan’s presidency in 2020.
Taiwan’s industrial transformation did change the cross-strait political landscape. However, it has not changed the Taiwanese subcontractors’ dilemma, which remains embedded in cross-strait economic interdependence under the competition of supercapitalism. Recently, Taiwanese business groups such as Foxconn are aggressively trying to either vertically integrate or upgrade by buying technology from certain foreign companies. In effect, the ‘salmon returning home’ policy was not influential until these business groups were seriously impacted by the recent US–China trade war. Business groups have not only downsized in China and returned to Taiwan, they have invested in upstream companies in the United States, Japan, and Korea and downstream factories in India and Southeast Asia. Taiwanese capital flight from China also reflects how vulnerable these firms are following fallout from the 2008 financial crisis and the vicissitudes of geopolitics, such as the ongoing US–China trade conflict. After an era of making easy money from the ‘world factory’ in China, for most Taiwanese business groups, the real challenge of upgrading is just around the corner.
Thung-Hong Lin
is an Associate Research Fellow and Joint Appointment Associate Professor at the Institute of Sociology, Academia Sinica, and National Tsinghua University. He is interested in social stratification, the sociology of disasters, and labour studies. He graduated from the Hong Kong University of Science and Technology and is an active participant in social movements in Taiwan, Hong Kong, and China. He is also the author of The Generation of Collapse (2011), a bestselling book on social critique in Taiwan.
Bowei Hu
is a PhD student in sociology at the University of California, Los Angeles. His research interests include social stratification and inequality, political economy, and organizations. He is working on a project about financialisation and income inequality.
References
Amsden Alice (1989) Asia’s Next Giant: South Korea and Late Industrialization, New York and Oxford: Oxford University Press.
Apple Daily (2012) 不能沒92共識，王雪紅挺馬 [‘Taiwan cannot live without 92 consensus’, Cher Wang stands by Ma], 14 January. Retrieved 30 May 2018 from https://tw.appledaily.com/headline/daily/20120114/33959958.
Babb Sarah (2005) ‘The social consequences of structural adjustment: Recent evidence and current debates’, Annual Review of Sociology 31: 199–222.
Baran Paul A. and Sweezy Paul M. (1966) Monopoly Capital: An Essay on the American Economic and Social Order, New York: Monthly Review.
bbc News (2017) ‘Trump takes credit for Foxconn’s “incredible investment”’, 27 July. Retrieved 17 August 2017 from http://www.bbc.com/news/business-40732035.
Block Fred , and Evans Peter (2005) ‘The state and the economy’, in Smelser Neil J. and Swedberg Richard (eds), The Handbook of Economic Sociology, New York: Princeton University, 505–526.
Burawoy Michael (1985) The Politics of Production: Factory Regimes under Capitalism and Socialism, London: Verso.
Burawoy Michael (2005) 製造甘願–壟斷資本主義勞動過程的歷史變遷, trans. Thung-Hong Lin et al., Taipei: Socio Publish. Chinese translation of Manufacturing Consent: Changes in the Labor Process under Monopoly Capitalism, Chicago and London: University of Chicago Press, 1979.
Bureau van Dijk (2018) ‘Top 2500 companies’. Retrieved 17 December 2018 from the Orbis database.
Castells Manuel (2000) The Rise of the Network Society, New York: Basil Blackwell.
Chandler Alfred D. Jr. (1977) The Visible Hand: The Managerial Revolution in American Business, Cambridge, MA: Belknap Press.
Chandler Alfred D. Jr. (1990) Scale and Scope: The Dynamics of Industrial Capitalism, Cambridge, MA: Belknap Press.
China Credit Information Service (ccis) (2016) ‘Top 5000 Taiwanese Business Ranking Database.’ Retrieved 6 August 2016 from https://www.credit.com.tw/newweb/DB/index_eng.cfm?Fuseaction=intro.
Chu Wan-wen and Amsden Alice H. (2003) Beyond Late Development: Taiwan’s Upgrading Policies, Cambridge, MA: mit Press.
Coase Ronald H. (1992) ‘The institutional structure of production’, American Economic Review 82(4): 713–719.
CommonWealth Magazine (2016) 2000 大臺灣企業調查 [The top-2000 Taiwanese company survey]. Retrieved 6 August 2016 from http://www.cw.com.tw/subchannel.action?idSubChannel=51#.
CommonWealth Magazine (2018) 兩岸三地1000大臺灣企業調查 [The top-1000 company of Great China region survey]. Retrieved 26 May 2018 from https://www.cw.com.tw/article/article.action?id=5090026.
Daepp Madeleine ; Hamilton Marcus J. ; West Geoffrey B. ; and Bettencourt Luis M. A. (2015) ‘The mortality of companies’, Journal of The Royal Society Interface 12(106). Retrieved 1 March 2019 from https://royalsocietypublishing.org/doi/full/10.1098/rsif.2015.0120.
Deyo Frederic C. (ed) (1987) The Political Economy of the New Asian Industrialism, Ithaca and London: Cornell University Press.
Fortune (2015) ‘Fortune Global 500’. Retrieved 12 October 2015 from http://fortune.com/global500/.
Friedman David (1988) The Misunderstanding Miracle: Industrial Development and Political Change in Japan, Ithaca, NY: Cornell University Press.
Gereffi Gary (1994) ‘The organization of buyer-driven global commodity chains: How U.S. retailers shape overseas production networks’, in Gereffi Gary and Korzeniewicz Miguel (eds), Commodity Chains and Global Capitalism, Westport, CT: Praeger, 95–122.
Gereffi Gary and Wyman Donald L. (ed) (1990) Manufacturing Miracles: Paths of Industrialization in Latin America and East Asia, Princeton, NJ: Princeton University Press.
Gereffi Gary ; Humphrey John ; and Sturgeon Timothy (2005) ‘The governance of global value chains’, Review of International Political Economy 12(1): 78–104.
Gilson Ronald J. ; Sabel Charles F. ; and Scott Robert E. (2009) ‘Contracting for innovation: Vertical disintegration and interfirm’, Columbia Law Review 109(3): 431–502.
Granovetter Mark (1985) ‘Economic action and social structure: The problem of embeddedness’, American Journal of Sociology 91(3): 481–510.
Haggard Stephan (1990) Pathways from the Periphery: The Politics of Growth in Newly Industrializing Countries, Ithaca, NY: Cornell University Press.
Hamilton Gary G. (1997) ‘Organization and market processes in Taiwan’s capitalist economy’, in Marco Orrù, Nicole Woosley Biggart, and Gary Hamilton (eds), The Economic Organization of East Asian Capitalism, Thousand Oaks, CA: Sage Publications, 237–293.
Hamilton Gary G. and Biggart Nicole Woosley (1988) ‘Market, culture, and authority: A comparative analysis of management and organization in the Far East’, American Journal of Sociology 94: S52–S94.
Hamilton Gary G. and Kao Cheng-shu (1987) ‘Max Weber and the analysis of East Asian industrialization’, International Sociology 2: 289–300.
Ho Ming-sho and Lin Thung-hong (2015) ‘Karl Polanyi in Taipei: Taiwan’s Sunflower Movement as the self-protection of a society’, paper presented at 中國效應工作坊：太陽花運動與中國效應 [The Workshop of China Effect: Taiwan’s Sunflower Movement and China Effect], Institute of Sociology, Academia Sinica, Taipei, 26 October.
Hsieh Michelle F. (2011) ‘Similar opportunities, different responses: Explaining divergent patterns of development between Taiwan and South Korea’, International Sociology 26(3): 364–391.
Hsieh Michelle F. (2014) ‘Hollowing out or sustaining? Taiwan’s sme network-based production system’, Taiwanese Sociology 28: 149–191.
Hsieh Michelle F. (2015) ‘Learning by manufacturing parts: Explaining technological change in Taiwan’s decentralized industrialization’, East Asian Science, Technology and Society 9(4): 331–358.
Johnson Chalmers (1982) miti and the Japanese Miracle, Stanford, CA: Stanford University Press.
Lamoreaux Naomi R. ; Raff Daniel M. G. ; and Temin Peter (2003) ‘Beyond markets and hierarchies: Toward a new synthesis of American business history’, The American Historical Review 108(2): 404–433.
Langlois Richard N. (2003) ‘The vanishing hand: the changing dynamics of industrial capitalism’, Industrial and Corporate Change 12(2): 351–385.
Langlois Richard N. (2004) ‘Chandler in a larger frame: Markets, transaction costs, and organizational form in history’, Enterprise and Society 5(3): 355–375.
Lee Zong-rong , and Hsiao Hsin-huang Michael (2014) ‘Taiwan: sme-oriented capitalism in transition’, in Witt Michael A. and Redding Gordon (eds), The Oxford Handbook of Asian Business Systems, Oxford: Oxford University Press, 238–259.
Lin Thung-hong (2015) ‘The making of collapse: Taiwan in the 21st century’, Global Dialogue 5(4): 30–31.
Lin Thung-hong and Hu Bo-wei (2017) 進擊的巨人：台灣企業規模迅速成長的原因與後果 [Size matters: The causes and consequences of the expansion of Taiwanese business groups], in Zong-Rong Lee and Thung-Hong Lin (eds), 未竟的奇蹟：轉型中的台灣經濟與社會 [Unfinished Miracle: Taiwan’s Economy and Society in Transition], Taipei: Academia Sinica, 230–266.
Lin Thung-hong ; Lin Yi-ling ; and Tseng Wei-ling (2016) ‘Manufacturing suicide: The politics of a world factory’, Chinese Sociological Review 48(1): 1–32.
March James G. and Simon Herbert A. (1958) Organizations, New York: John Wiley and Sons.
Mathews John A. and Cho Dong-sung (2000) Tiger Technology: The Creation of a Semiconductor Industry in East Asia, Cambridge: Cambridge University Press.
National Development Council (2016) 加強推動台商回台投資方案 [The project of encouraging the returnee investment of Taiwanese businesspersons]. Retrieved 30 May 2018 from https://www.ndc.gov.tw/News_Content.aspx?n=9D32B61B1E56E558&sms=9D3CAFD318C60877&s=06B0C4946C5B5749.
O’Conner, James (1973) The Fiscal Crisis of the State, New Brunswick, NJ: Transaction Publishers.
Offe Claus (1984) Contradictions of Welfare States, Cambridge, MA: mit Press.
Piketty Thomas (2014) Capital in the Twenty-First Century, Cambridge, MA: Belknap Press.
Piore Michael J. and Sabel Charles F. (1984) The Second Industrial Divide, New York: Basic Books.
Reich Robert B. (2007) Supercapitalism: The Transformation of Business, Democracy, and Everyday Life, New York: Alfred A. Knopf.
Rodrik Dani (2011) The Globalization Paradox: Democracy and the Future of the World Economy, New York and London: W.W. Norton.
Newsroom Samsung (2019) ‘About us: Fast facts’. Retrieved 14 January 2019 from https://news.samsung.com/global/fast-facts.
Schumpeter Joseph A. (1994) Capitalism, Socialism and Democracy, 5th ed., London: Routledge. First published 1942.
Sharkey William W. (1982) The Theory of Natural Monopoly, Cambridge: Cambridge University Press.
Shieh Gwo-shyong (1992) ‘Boss’ Island: The Subcontracting Network and Micro-Entrepreneurship in Taiwan’s Development, New York: Peter Lang.
Uzzi Brian (1999) ‘Embeddedness in the making of financial capital: How social relations and networks benefit firms seeking financing’, American Sociological Review 64: 481–505.
Wade Robert (2004) Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization, rev. ed., Princeton, NJ and Oxford: Princeton University Press. First published 1990.
Wall Street Journal (2019) ‘Foxconn Technology Co. Ltd., 2354 (Taiwan)’. Retrieved 14 January 2019 from https://quotes.wsj.com/TW/XTAI/2354/financials?fbclid=IwAR3Xaa3IL6CzUMw9zFA9VpPG6VPONHnycSksaLesa0sYY6Eh14Be6BwKZNQ.
Williamson Oliver E. (1981) ‘The economics of organization: The transaction cost approach’, American Journal of Sociology 87(3): 548–577.
Wu Nai-teh (1987) The Politics of a Regime Patronage System, PhD thesis, University of Chicago.
Wu Yong-ping (2004) ‘Rethinking the Taiwanese developmental state’, The China Quarterly 177: 91–114.
The study presented here follows on from Lin & Hu (2017) but makes use of new statistical data and modelling, and elaborates our theory with new empirical findings.
With some exceptions, for example Lin, Lin, and Tseng (2016).
These policies include the 2010 Economic Comprehensive Framework Agreement, the ‘salmon returning home’ (guiyu fanxiang) policy formally applied since 2013, and the Cross-Strait Service Trade Agreement (cssta) in 2014 (Ho & Lin, 2015).
The most important incentive of the project was the expanded admission of migrant workers from Southeast Asia and lower interest rates on Taiwanese bank loans, but these failed to attract returnees and were ended at the end of 2014 (National Development Council, 2016).
For example, in the last month of the presidential election in 2012, the executives of big Taiwanese groups, such as Foxconn, Evergreen, and htc (Formosa Plastics Group), openly echoed the Chinese Communist Party’s policy to sustain the kmt’s ‘1992 Consensus’ ( Apple Daily, 2012).
Rex How (郝明義) is a famous Taiwanese publisher who openly supported the Sunflower Movement because of concerns that opening the publishing and mass media industries to Chinese investment may damage publishing and free speech in Taiwan.
