East Asian Varieties of Capitalism and Socio-Economic Inequality: South Korea and Hong Kong Compared

Abstract This article examines the deepening of socio-economic inequality manifested as dualisation in South Korea and deregulation in Hong Kong. It explains the extreme manifestations of inequality by reference to the nature of economic co-ordination, organised labour power and societal corporatism, and the politics of democratisation. It finds that South Korea’s legacy of state-orchestrated co-ordinated market economy favoured the retention of a manufacturing core despite globalisation. However, the neo-liberal inclinations of the state and big business groups led to the marginalisation of less privileged firms and workers. These inclinations reinforced the defensiveness of the small but strong labour organisations, preventing effective societal corporatism. These phenomena are fully understandable only with reference to the country’s conservative democratisation that divided the liberal and left political forces. The geo-political legacy also induced identity conflicts that overshadowed socio-economic issues. By contrast, Hong Kong’s liberal market economy coupled with the absence of security concern had allowed a radical de-industrialisation in the 1990s and wholesale casualisation in the 2000s. Divided by identity politics, organised labour was unable to challenge the trend. The same conflicts also led pro-China labour organisations to side with business interests in the post-1997 electoral autocracy and prevented the introduction of more fundamental reforms.

contrasts with the advanced Western capitalist economies where economic maturity was accompanied by a "golden age" of shared prosperity from the late 1950s to the mid-1970s.Furthermore, despite their apparent similarities, Hong Kong experienced a drastic and wholesale increase in income inequality since the 1990s; yet Korea underwent a less extreme increase that affected segments of the population more seriously than others.Put another way, just as the labour market of Hong Kong underwent deregulation, that of Korea saw a deepening of dualisation. 2 This article makes three inter-related arguments to account for these different scenarios.First, political-economic theories that examine globalisation and productive structures present useful starting points for explaining the East Asian economies' transformations, and in particular, the varieties of capitalism (VoC) approach (Hall and Soskice 2001).This approach scrutinises the level of employer co-ordination, and can usefully distinguish Korea's co-ordinated market economy (CME), albeit state-orchestrated, from Hong Kong's liberal market economy (LME), and explains Korea's higher level of egalitarianism before globalisation.However, given its reliance on path dependency to explain an economy's response to globalisation, the VoC approach has difficulty examining political actions or accounting for divergencies from the expected pathways of transition.
Second, this article will seek a critical integration with approaches that emphasise political-institutional factors, especially the VoC approach modified by Thelen (2012Thelen ( , 2014)), as a remedy to the aforementioned inadequacy of the political-economic theories.Thelen's modified-VoC approach draws attention to labour unions and corporatism. 3She delineates three pathways of labour market transitionderegulation, dualisation, and flexicuritisation 4 amongst twenty-first century LME and CME cases.Adopting this approach, it becomes possible to see that, just as the weakness of organised labour in Hong Kong has engendered deregulation, strong organised labour in the absence of societal corporatism in Korea has triggered dualisation at the expense of the non-unionised workers.Yet, again, Thelen's approach cannot explain the apparent extremity of deregulation and dualisation in the two cases.
Third, this article highlights the importance of geo-politics for Korea and Hong Kong's economic and political development and hence the need to situate the VoC approach and Thelen's valuable political-institutional analysis within pertinent global and regional dynamics.In doing so, it strives to explain more precisely why Korea and Hong Kong have adopted divergent strategies of globalisation-induced restructuring.Specifically, geopolitics has profoundly affected the East Asian state elites' security considerations in relation to China's economic rise in the post-Cold War era and, hence, their willingness to relocate production to the country.With keen concerns for national security, Korea has cautiously retained a significant albeit diminished core domestic manufacturing capacity (Kim and Kwon 2017).At the same time, its big business groups (BBGs) have sought to diversify their supply chains and markets.The continuing pursuit of CME-type, highquality manufacturing maintains the position of the labour aristocracy and preserves an uneasy inter-dependence between the BBGs and their (reduced) unionised core labour.In turn, the co-existence of an organised labour core with weak societal corporatism has accentuated dualisation, leaving the rest of the labour force to feel the full squeeze of neoliberalism.By contrast, the lack of security constraints reinforced the existing liberal institutional set-up (that is, weak employer co-ordination) in Hong Kong to engender a wholesale relocation of manufacturing activities to mainland China.The weakness of organised labour and absent societal corporatism generated a radical deregulation of the labour market.
At the same time, not only had geo-politics bolstered authoritarian domination during the Cold War era, but it also contributed to the consolidation of political and economic interests that continue to take effect in the process of democratisation.In the case of Korea, "conservative democratisation" has preserved the overarching positions of the BBGs and entrenched political parties, marginalised the emerging labour powers, and rendered it difficult both for the employers and unionised workers to strike a compromise, and for the state to compel vested interests towards co-operation.Furthermore, the background of decolonisation and the Cold War bequeathed Korea's democracy with legacy issues that competed with socio-economic ones in saliency.Moreover, the left was divided between opposition liberals with a stake in the former quasi-democracy and more radical forces based on the independent labour movement.Though removed from direct Cold War confrontations, Hong Kong's democratisation was stalled under colonial domination.Above all, politics in the territory has also polarised according to geo-political fault-lines, namely, pro-versus anti-China.With the introduction of electoral autocracy following its handover to China, the territory's pro-China politicians have aligned with business elites to thwart fundamental labour and welfare reform.Instead, they have settled for stopgap measures to gain the support of segments of the population and address the needs for social protection as they arise.Even though Korea's liberal-democratic system, by allowing for the rotation of power, appears more flexible than Hong Kong's hybrid system in responding to social dysfunctions, the East Asian cases still differ from the advanced Western capitalist economies where a symbiosis between democracy and advanced capitalism allegedly exists (Iversen and Soskice 2019).These explain the two cases' more extreme forms of dualisation and deregulation.Figure 1 encapsulates the two respective trajectories.
The following section will examine the political-economic theories and their shortcomings, discuss political-institutional theories to facilitate a critical integration, and introduce geo-political dynamics to complete the theoretical framework.The article will then explain the choice of the two cases and analyse them from three angles: employer co-ordination, organised labour and societal corporatism, and the political system and politics of democratisation.This article will conclude with a summary of the arguments and a reflection on the study's theoretical significance.

Intensification of Inequality in East Asia: Toward an Analytical Framework Centring on Korea and Hong Kong
East Asia's advanced economies have made impressive strides since their "miraculous" development between the 1960s and the 1980s.Subtle state interventions help local corporations to integrate advantageously into global value chains (see Kalinowski 2015;Chu 2021).Business elites also co-operate willingly in industrial upgrading, rather than act as a vested interest as in the "middle-income limbo" economies of Southeast Asia and Latin America (Doner and Schneider 2016, 620-625).Impressive income growth, however, has not been matched by shared prosperity as in the "golden age" of Western capitalism. 5For East Asia, high-income status has coincided with the worsening of inequality, stagnant incomes for the majority, contraction of stable employment, and underdeveloped social welfare.These shortcomings are apparent from comparisons with the advanced Western economies, including the highly liberal market ones (see Figures 2 and 3).
Such economic and social weaknesses cannot be adequately explained by theories of globalisation and changing productive structures.Arguments of financialisation aptly highlight activities (such as mergers and acquisitions) that tend to reduce employment and compress wages for the sake of extracting maximum short-term value (Davis and Kim 2015, 210).However, financialisation has limited explanatory value in these cases given the low level of financialisation as measured by the stock market/gross domestic product (GDP) ratio in Japan, Korea, and China (Kalinowski and Jang 2014, 503).
Analysts of global production networks have identified a tendency for some East Asian business corporations to "dis-embed" from their national economies while intensifying their offshore production activities, resulting in the transfer of manufacturing jobs to China and Southeast Asia where lower wages prevail (see, for example, Yeung 2016).Consequently, manufacturing employment is replaced by services sector employment where pay is polarised between a small elite (concentrated in finance, high technology, senior management, and legal services) and the low-paid "precariat" (Haggard 2016, 10).Although perceptive, this argument leaves open further questions regarding why some governments are more passive in the face of socially harmful offshoring and why some firms are more willing and able to discard their local supply chains.GDP, 1990GDP, -2015 Source: Data for Germany, Japan, South Korea, and the USA are from OECD (2021).For Taiwan, data are from DGBAS (2019).The data for Hong Kong are from HKCSD (various years).Singapore's data are from Government of Singapore (2022).Note: For Germany, Japan, South Korea, and the USA, public social spending includes spending on old age, survivors, incapacity related, health, family, active labour market program, unemployment, housing, and other social policy areas.The data for Hong Kong are calculated by adding the expenditure on housing, health, and social welfare drawn from the table on "Public expenditure by policy area group."The figure in 2000-2001 increased abruptly because housing expenditure was more than three times that of 1991-1992.The figures are not directly comparable to those of the OECD.Singapore's data refer to all "social development" items in the government's total expenditure.
The issue of diversity in the trajectories of labour market transition and inequality can be partially explained by the VoC approach and "growth regime" analysis.Both underscore the path dependency effects of productive structures.The original VoC approach divides advanced capitalism between CME and LME, separated by their co-ordination mechanisms (market or non-market) in five complementary institutional domains (industrial relations, vocational training and education, corporate governance, finance, inter-firm relations) (Hall and Soskice 2001, 6-7).The complementary nature of these institutions shapes employers' responses to globalisation in path-dependent ways.The tighter the level of employer co-ordination at the start of globalisation, the less inegalitarian the form of globalisation will take.As employer co-ordination is characterised by centralised wage bargaining and wage restraint, this generates financial resources for employers to invest in the ongoing development of industry-specific or firm-specific skills that boost continuous productivity growth in the lead manufacturing sector.It follows that a system of long job tenure and strong employment protection will enable both employers and employees to realise the long-term benefits of investment in training (Hall and Soskice 2001, 24-26;Thelen 2014, 12).However, even as the VoC approach has accurately postulated globalisation's effect in deepening liberalisation and aggravating inequality among the LMEs such as the USA and the UK, some leading CMEsnotably Germany and Japanhave become inegalitarian in ways not entirely explicable within the original VoC approach.
Hassel and Palier's (2021, 42-50) more wide-ranging typologies of "growth regime" have covered the inegalitarian turn of some CMEs. 6For these authors, policymakers are inclined towards welfare policiesdefined broadly to include labour markets and educationto preserve and reinforce existing productive specialisations.Dualisation is seen in Germany as resulting from the attempt to preserve the competitiveness of the high-quality manufacturing export regime by partial labour market reforms that reduce labour costs outside the manufacturing core.However, this approach cannot explain why dualisation is more severe in Korea than Germany (and Japan) given the predominance of quality manufacturing in all three economies.Above all, an argument that falls back on the logic of path dependency leaves unaddressed the political choices and social forces that underlie the policymakers' decisions.
Against this backdrop, this study seeks to supplement the productive structure-centric explanations with an analysis of political-institutional factors.As Thelen (2014, 139) points out, the original VoC thesis conflated the persistence of a high level of employer co-ordination with a high level of equality leaving unexplained variations between CMEs once thought to be very similar.She proposes the incorporation of two politicalinstitutional factors to analyse such divergencies.The first factor is the relative power of organised labour and employers and the nature of their relationship as derived from Korpi's (2006) "power resource" approach.The second, drawing from corporatist theory, is state effectiveness in brokering political settlements that reshape labour-employer relations that may curb the marginalisation of peripheral firms and workers (see Thelen and Kume 2006;Martin and Thelen 2007).The question is then whether organised labour power and societal corporatism reinforce or modify the legacy of inherited productive structures.The presence of strongly organised labour can deter a deeper neo-liberalisation but its existence in the absence of societal corporatism can also trigger a form of partial neo-liberalisation highly disadvantageous to non-unionised "outsiders."These politicalinstitutional factors underpin Thelen's explanation of the divergent trajectories of two pairs of CMEs: Germany-Netherlands and Sweden-Denmark.
Thelen's modified VoC approach, though helpful in delineating the divergent pathways of deepening inequality, does not adequately account for the extent of dualisation and deregulation when considering Korea and Hong Kong.To this end, this study proposes to situate the original VoC approach and Thelen's modification within an analysis of global and regional political dynamics that have exerted a wide-ranging impact on East Asia.Emerging as a "divided nation" or a colony from World War II and industrialised through the Cold War years, it would not be an exaggeration to postulate that geo-politics reverberates in every step of economic and political institution building in the two cases.Geo-politics can also be expected to mediate the decisions of state and social actors on globalisation-induced restructuring as well as cut through labour relations and democratic politics.As such, this article will examine the extent to which the political institutions and divides bequeathed by Cold War geopolitics and decolonisation have affected the political and economic elites' readiness to incorporate labour as well as labour's opportunities for political organisation.Have these political trajectories lent weight to continuing exclusion and diverted political energies away from resolving socio-economic issues?In highlighting East Asia's complex social and political relations, the analysis will also reflect on Iversen and Soskice's (2019) argument that, given capital's relative immobility and the need for electoral support from the middle class (and its aspirants), there exists a symbiosis between advanced capitalism and advanced democracy.
In short, this article proposes to critically integrate the original VoC approach, Thelen's modification, and geo-politics, and examine the cases with three analytical foci.The first will centre on the economic institutions and changes in employer co-ordination.The second will focus on political-institutional factors or organised labour and societal corporatism.The third will address democratisation and its effect on socioeconomic issues.
This study has chosen to focus on Korea and Hong Kong because, even as they embarked on industrialisation at about the same time, the two cases display divergent trends in the deepening of socio-economic inequality (see Figures 2 and 3), differ markedly in the level of market co-ordination, and exhibit most clearly the two distinct developmental pathways to high-income status, namely, Korea as manufacturing-centric versus Hong Kong's services-orientation (see Figure 4).The two cases also illustrate how geopolitics has triggered divergent political institutionalisation: Korea's conservative democratisation versus Hong Kong's electoral autocracy.Although they make rather unusual choices for comparing the East Asian cases, their contrast could yield insights into what might be considered delimiting situations.Table 1 summarises their comparability with the two advanced Western capitalist varieties identified by Thelen (2014).The next section will examine them in detail using the approach outlined above.

Legacy of Employer Co-ordination
During the era of Korea's "development miracle" (c. 1965-1990), the authoritarian state introduced CME-type institutions to prevent "over-competition," direct investment towards promising sectors, exercise wage restraint, and facilitate labour-management cooperation.Compared to classic CMEs like Germany and Japan, however, co-ordination in Korea was less deeply ingrained, and more reliant on state compulsion.Thus, it was more vulnerable to unravelling under the conditions of political and economic transition during 1987-1997 (Kang 2010, 534-537).This resulted not in a transition to a wholly liberalised labour market, but in further dualisation, as in Germany and Japan.
The 1997-1998 economic crisis led to the introduction of unprecedented labour market reforms.The principal reform measures were the legalisation of layoffs for "business reasons" and the easing of "atypical" employment.These measures enabled employers to replace regular workers with non-regular workers paid at lower rates, without fringe benefits or security of employment (Fleckenstein and Lee 2019, 459).Regular employment eventually recovered after the crisis, reaching an all-time high in 2015.Figure 5 shows that regulars constituted 65.5% of waged workers and 48.5% of all workers, including "non-waged" self-employed and family workers in 2015.
Dualisation also deepened after the financial crisis.The share of non-waged selfemployed and family workers in Korea (25.9%) remained higher than that of other advanced market economies: 15.2% in the UK, 11.5% in Japan, 10.8% in Germany, and 6.5% in the US (KLI 2016a).Among wage workers, the pay gap widened at the expense of non-regular workers, as reflected in the increase of the Gini index (see Figure 6).Furthermore, whereas regular workers' coverage by employment insurance increased from 82.8% in 2005 to 87.8% in 2015, the coverage of non-regular workers decreased from 17.3% to 12.1% during the same period.
Remuneration and job stability have also bifurcated according to enterprise size.Between 1995 and 2015, the share of employees working at large enterprises (LEs, with 500-999 workers) fell from 18.5% to 14.3%.The share for very large enterprises (VLEs, with 1,000-plus workers) declined from 9.2% to 6.8%.The pay gap with small-and-medium enterprises (SMEs), however, has been increasing (see Figure 7).Job stability also favoured  LEs.For LEs, tenure increased from 7.3 to 10.4 years between 1995 and 2015, compared with the overall average increase from 5.7 to 6.5 years over the corresponding period.This pattern of dualisation reflects the partial unravelling of employer co-ordination.First, state-orchestrated employer co-ordination in Korea was noted for its weakness in   III-4; (2017), Tables III-3 and III-4.the domain of labour-management relations.During the "miracle" era, employers relied more on state coercion than voluntary co-operation to check wage increases.This started to unravel in 1987 as democratisation generated opportunities for labour strikes.The number of workdays lost to strike action escalated from 72,000 in 1986 to 6.94 million in 19876.94 million in , averaged at 5.4 million per annum between 19886.94 million in and 19906.94 million in , and declined gradually to 393,000 in 19956.94 million in (KLI 1999, 203), 203).Unable to contain wages, BBGs responded by rolling back the job stability and fringe benefits introduced to retain high skilled labour and facilitate industrial peace during the 1970s-1980s.
Co-ordination, however, did not completely unravel as workers in the BBGs remained relatively insulated.Although Korean BBGs and their unionised workforces failed to develop mutual trust like their Japanese counterparts, their mass production structures created certain forms of interdependency that could not be easily discarded (see Song 2012).One form was continuous investment in firm-specific skills through vocational training.The "residual expenditure" category of labour cost, which included expenditure on vocational training, remained high for VLEs in absolute and relative terms (see Figure 8). 7 To realise the benefits from investments in vocational training, BBGs needed to sustain the loyalty of their workers through the continual provision of job security and corporate welfare benefits (Chung and Lee 2015, 21-22).Much as they wished to reduce their core labour forces, Korean BBGs could not simply relocate their core manufacturing facilities and supply chains overseas.

Power Resources of Labour and Societal Corporatism
Given their weaker reciprocal relationship with their core labour forces, Korean BBGs were more inclined towards neo-liberalism than their counterparts in CMEs such as Japan and Germany.Correspondingly, their largely unionised workforces were more defensive.The BBGs' and unionised workers' conflicting impulses squeezed those who worked for the BBG's sub-contractors as well as non-regular workers within the BBGs themselves.The latter constituted as much as 38% of the top 10 BBGs' workforces in 2016.For the 11th-53rd ranked BBGs, the composition was 45-47% (KLSI 2016, 9).The nature of these relationships highlights the importance of labour's power resources in shaping Korea's more extreme path of dualisation in comparison with other CMEs.
Union density in Korea peaked at 17.2% in 1990 and declined thereafter.Membership growth resumed from the mid-2000s, but density stabilised at around 10%.This was similar to the US (11.1%) but well below Germany (19%) and Japan (17.4%) and the UK (24.7%) in 2015 (JILPT 2018, 225).The number of workdays lost to strike action was also  falling by the 2010s (see Figure 9).These trends show labour's power resources to be diminishing.
These broad indicators conceal the variations of labour strength.Among regular workers, the unionisation rate was about 20% and, for VLEs, it was as high as 73% in 2015 (KEF 2017, 36).For unionised workers, the agreed pay norms have always exceeded average real wage increases for regular workers in the best paid manufacturing sector (see Figure 10).Among the two main national labour confederations, the constituents of the smaller Korean Confederation of Trade Unions (KCTU) had a more concentrated membership, especially in heavy export industries (automotive, shipbuilding, steel) and the public sector.By 2015, it had 605,000 members in 302 enterprise branches.This included the 57,000-strong Hyundai Motors Workers Union, a member of the 146,000-strong Korean Metal Workers' Union (the mainstay of the KCTU).By contrast, the 820,000 members of the Federation of Korean Trade Union (FKTU) were dispersed across 5,749 enterprise branches.At the other extreme, less than 1% of non-regular workers were organised.
The uneven distribution of labour strength had implications for labour market dualisation.It was much more difficult for employers to impose neo-liberalism on organised workers.For example, Kim (2016, 46-47) has observed how union representatives sought to limit companies' direct hiring of non-regular labour but acquiesced to indirect (agency) hiring, especially for demanding jobs.They did this to prevent employers from directly hiring non-regular workers that could substitute for their members.To reduce costs, the large employers, and especially the BBGs, turned to externalisation where "outside" workers are hired through agency work contracts or sub-contracting.In particular, prime contractorsoften unionised BBG affiliatescould use their exclusive relationships to force sub-contractors into accepting low prices, leading to low wages and long working hours  (Yun 2011, 536-567).Focused on advancing the pay and working conditions of their members, powerful enterprise-level labour unions were less concerned with employers transferring the costs onto non-unionised workers, situated mainly in SMEs, resulting in the worsening of labour market dualisation.In the long term, however, externalisation diminished the effectiveness of industrial action at core enterprises, thereby reducing the unions' bargaining power as well as the appeal of union membership (see Nho and Kim 2018).
Against this backdrop, employers blamed self-interested labour unions as the driver of externalisation and labour market dualisation.They called for the easing of employment termination for regular workers to "vitalise" the labour market (KEF 2016, 15), so as to create opportunities for those most disadvantaged by dualisation.However, there were also limits to this simple logic that traced labour market dualisation to powerful unionised insiders.Despite earning high profits, BBGs were reluctant to boost the wages and productivity of their sub-contractors by making investment.The BBGs also retained very high cash reserves.For example, in 2015, 96 publicly listed companies of the top 10 BBGs held cash reserves (including invested assets) of over 500 trillion won (US$451.49billion) (Hankyoreh, March 24, 2015).In their campaigns for public social protection and improvement of the rights of non-regular workers, the labour confederations (especially the KCTU) revealed their commitment to solidarity born out of ideology and strategic self-interest (Fleckenstein and Lee 2019, 459-460).Union federations at the industrial level have also made gestures of solidarity on behalf of non-regular workers.For example, the Korean Metal Workers' Union proposed to divert 250 billion won (US$219.9 million) of regular workers' wage claims to support non-regular workers of sub-contracting companies (Hankyoreh, June 20, 2017).
The Korean state was unable to overcome this impasse with corporatist initiatives.In part, this reflected the power imbalances within the KCTU described above.The roots of the KCTU's reluctance could also be traced to the experience of the 1998 tripartite agreement.Agreed on February 9, 1998, this was meant to be a burden-sharing response to the financial crisis.The leaderships of the national labour confederations accepted wide-ranging labour market reforms in exchange for social guarantees and new political rights.In the same month, however, the special conference of the KCTU delegates rejected the agreement and dismissed the executive signatories (Baccaro and Lim 2007, 33).In part, the withdrawal of the KCTU reflected the imbalances of union power in favour of the BBG-based and public sector labour unions that were more inclined towards resistance.By contrast, weaker unions of the SMEs were more inclined towards negotiated restructuring as the way of protecting their members' interests (Baccaro and Lim 2007, 32-33).Seeking to avert internal division and disappointed with the social consequences of restructuring, subsequent KCTU leaders boycotted all initiatives.The absence of the KCTU meant that co-operation could not reach into the most influential sections of the union membership.
However, the failure of societal corporatism went beyond the internal power balances within the KCTU.It also revealed persistent government deficiencies, notably persistence of top-down state tradition (as well as opportunistic use of social dialogue to legitimise official policies) (Kim and Ahn 2018;Chang 2018). 8These tendencies worked against the development of genuine social partnership.They were evident from the pattern of cooperation and withdrawal involving the FKTU.Given its membership concentration in SMEs, the FKTU unions were more vulnerable to neo-liberal policies and so had more incentive for making an alliance with the government.Hence, the FKTU tended to support government initiatives and forged tripartite agreements with governments both liberal (1998) and conservative (2009 and 2015).Nevertheless, FKTU participation was also relatively short-lived.Its support for the 1998 social pact lasted just 13 months.The FKTU left claiming that the liberal government did not apportion the costs of restructuring fairly.In 2009, it left the social pact after seven months in a dispute with the conservative government over the company payment of union officials, in which it made cause with the KCTU (Hankyoreh, October 16, 2009).
The collapse of the 2015 social pact after four months was emblematic of the government deficiencies identified above.As the conservative candidate in 2012, President Park Geun-Hye acknowledged the seriousness of labour market dualism and pledged to increase, maintain, and upgrade jobs as well as re-start social dialogue (Chang and Park 2017, 3).Social dialogue resulted in the Tripartite Agreement on Structural Reforms of the Labour Market of September 2015 between the government, FKTU, and Korea Employers Federation (ESDC 2015).The agreement set out eight key objectives beneficial to all stakeholders in the labour market (KMOEL 2015). 9By January 2016, the FKTU had withdrawn from the agreement citing the absence of government consultation and onesided implementation of controversial measures relating to dismissal for under-performance and misconduct, as well as extension of the period of use of fixed-term workers (KLF 2016).Ruling party politicians responded by denouncing the labour confederations as "labour aristocrats" unconcerned with the plight of vulnerable workers (Yun 2017, 501).

Nature of the Political System
Leading explanations of inclusive growth in "rich democracies" emphasise: the "power resources" of the left (Korpi 2006); the capacity of the state to compel and persuade vested interests towards "flexicurity" (Thelen 2014, 31-32); and the effect of political competition in creating competitive markets (Iversen and Soskice 2019, 53-101).In the case of Korea, however, the inclusionary effects identified in these accounts are constrained by the political institutions and political divides that sprang from its pathway of conservative democratisation.The three types of political constraint were: the division among the leftwing forces; the incapacity of the state to compel or persuade the vested interests towards co-operation; and the inability of competitive politics to translate into competitive economics.This section will show how each of these constraints emerged out of Korea's democratisation path and highlight their socially exclusionary effects.
"Conservative democratisation" refers to the introduction of competitive elections without jeopardising the vested interests formed under the authoritarian regime.This means protection of former authoritarian power holders and avoidance of radical redistributive measures inimical to business and middle-class interests.This was Cumings' (1989) pessimistic prognosis made soon after Korea's democratisation of 1987.This prognosis, however, was not so apparent during the first decade of democratisation (1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997).In contrast to other newly democratised countries, Korea inherited relative equality ("growth with equity") from authoritarian rule.Against the background of BBG-led growth, in promoting democratisation, workers successfully mobilised for redistribution.The nascent KCTU gained a seat at the tripartite forum in 1996.Politically, the double rotations of power (1997 and 2007) showed that Korea had passed the standard tests of democratic "consolidation" such as popular acceptance of democratic rules and "learning to lose" on the part of authoritarian incumbents (see Diamond 1999;Reidl et al. 2020, 316-317).Socially, the liberal government (1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008) responded to the 1997-1998 economic crisis by introducing labour market liberalisation alongside social welfare measures.This could be interpreted as a first step towards "flexicurity" Korean-style (Fleckenstein and Lee 2017, 172).For Cho (2012), the intensified political conflicts during the liberal decade signified the beginning of a transition from political de-monopolisation to economic and social de-monopolisation.
Nevertheless, social outcomes became more exclusionary after the 1997-1998 crisis.They set the foundations for the labour market dualism of the 2010s.These outcomes could be traced to the conservative path of democratisation.For one, Korea lacked the solidarity of left forces that enabled European social democratic parties to introduce redistributive policies and, later, "flexicurity."Democratisation divided Korea's "left" between the existing liberal opposition parties and the grassroots democratic activists associated with the independent labour movement.Grassroots activists played an active role in democratic mobilisation (Bellin 2000, 199-202).However, they were excluded from the subsequent democratic settlement.The existing liberal and conservative parties (including authoritarians) redesigned the electoral system in 1987 in their own interest.They opened it sufficiently to enable competition among themselves but closed it off to other potential entrants (Mobrand 2018a, 617).For example, they retained authoritarian-era laws to restrict the organisation of parties at the local level, to the detriment of labour-based parties adept at grassroots organisation (Mobrand 2018a, 609).Many liberal opposition parties grew out of the Democratic Party that was formed in 1955.Frequently an ally of authoritarianism during the 1950s, it became the acceptable opposition to the militarybased regimes   (Mobrand 2018a, 614-615).As such the two post-1987 successors of the Democratic Party (led by Kim Young-Sam and Kim Dae-Jung) secured political advantages that they sought to preserve and extend under democracy.They did not need the emergence of another opposition party on the left.
The left divide led each side to thwart the other's agenda.Inclusive European pathways such as "flexicurity" that interested liberal Korean politicians depended on government securing co-operation from labour unions and employers.The KCTU first entered into social dialogue in 1996, but that dialogue collapsed when the conservative government attempted to introduce flexibility measures unilaterally.However, the liberal government formed in February 1998 failed to secure KCTU support for its tripartite agreement for burden sharing (see previous section).The KCTU rejection of the agreement reflected more than the power imbalances between sections of the membershipthose in BBGs versus those in SMEsmentioned above.It occurred against a background of distrust between the KCTU and the new liberal government that could be traced back to the 1987 democratic settlement.In the 1990s, the KCTU became home to grassroots radical activists that played a leading role in the democratisation movement but who were subsequently excluded from the design of the new democratic order by the mainstream conservative and liberal parties (Mobrand 2018b).The KCTU's subsequent boycott of social dialogue was driven not only by the unacceptable levels of unemployment and casualisation, but also by government under-delivery on its promises, including the slow implementation of unionisation rights for teachers and unemployed workers, and inadequate levels of social welfare expenditure levels (Han, Jang, and Kim 2010, 300).Without the participation of the KCTU as the most powerful labour confederation, the Koreanstyle "flexicurity" agenda could not advance.
The divisions with the liberal government led the KCTU to support the formation of a more conventionally socialist electoral alternative, the Democratic Labour Party (DLP).In its first electoral test in 2004, the DLP captured 10 national assembly seats by proportional representation (out of 300).However, the party failed to build on this launchpad and dropped back to five seats (all proportional representation) in the 2008 election.One factor in this decline was the division of the party into competing ideological factions.These divisions originated from the ideological divides and historical tactics of left-wing antiauthoritarian groups during the 1980s.Contrasting the Korean left with Brazil's Labour Party, Lee (2016, 232-233) observed that while they managed to develop influence in the workplace in opposition to authoritarian rule, left-wing activists in Korea failed to build links with local community associations that were integral to workers' lives.This would later hamper DLP efforts to build the local support bases needed to capture geographical constituencies.This also helps to explain the loss of electoral momentum after 2004.On top of internal divisions, left alternatives continued to face institutional constraints arising out of the 1987 political settlement.For example, the liberal government tightened restrictions on local participation by prohibiting local party branches in 2004 (Mobrand 2018a, 609).The left-wing regrouped as the United Progressive Party (UPP) and captured 13 seats in the 2012 general election.However, it was constrained by the pro-North Korea ideology (chuch'e sasang) of some of its legislators and by the institutional environment.These factors enabled the conservative government to invoke the authoritarian-era party cancellation law to have the UPP dissolved in 2013 on the grounds of pro-North Korean subversive activity, even though the UPP possessed no actual coercive or organisational capacity to overthrow the state.The dissolution was also supported by most opposition liberal party legislators in the national assembly from whom one might have expected more critical challenge to the conservatives and the intelligence agencies (Mobrand 2018a, 614). 10The broader point is the main parties' acceptance of authoritarian-era restrictions that stifle smaller parties of the left.Even though those parties were small, they reflected labour activists with considerable influence in the KCTU.
Conservative democratisation also facilitated the growing economic power of the BBGs, and by extension, the path of jobless growth.The dominant political parties (conservative and liberal) under democracy shared the premise that economic liberalisation would correct the distortions created by authoritarianism, including economic monopolisation.The economic agenda of President Kim Dae-Jung reflected strong orthodox liberal market orientation.It believed that the financial crisis had been caused by distorted financial liberalisation (due to the authoritarian legacy) rather than liberalisation per se.Its over-optimism regarding the curative powers of liberalisation was exemplified by the unintended consequences of its signature reform requiring deep reductions of the BBGs' debt ratios.However, increased stock market reliance actually resulted in greater economic concentration favouring those BBGs that controlled the most lucrative export industries for example, Samsung electronics, Hyundai Motors, and SK Hynix) while weaker BBGs and SMEs lost ground.Of the top 30 BBGs, the top four accounted for 52.7% of assets and 69.4% of profits in 2017.The corresponding figures for 2011 were 49. 1 and 62.3% respectively (OECD 2018, 9-10).Increased stock market dependence, however, did not transform corporate governance because the founding families found ways to transfer control to successor generations by means of share manipulation as illustrated by the case of Samsung (Chang, Kim, and Park 2012, 79-83).These developments gave BBG founding families immense political influence since the fate of the economy would turn on the individual decisions taken by a few business tycoons.This meant the capacity for collective political action identified by Iversen and Soskice (2019) was less important in Korea than in other "rich democracies." What about the effect of political competition in checking the power of capital?For example, Iversen and Soskice (2019) argue that parties competing for power in "rich democracies" must satisfy key middle-class and aspiring middle-class voters by providing public goods and regulations (including a competitive economic environment) in order to nurture dynamic capitalism.As seen above, under conservative democracy, the dominant parties accepted the premises of the economic liberalisation agenda.In place of competition on economic issues, they motivated their supporters using historical legacy issues.During the 1990s, rival former democratic leaders Kim Young-Sam (conservative) and Kim Dae-Jung (liberal) both appealed to regional sentiment to mobilise their core supporters.While regional sentiment declined with the ending of the era of the Kims in 2002, the divide over legacy issues continued to intensify.Redistribution was an issue in the conservative-liberal divide, but within the parameters of both sides' neo-liberal premises.More potent were the divides over issues relating to North Korea, the US alliance and the achievements of the authoritarian past.This reached a climax at the 2012 presidential election when the conservatives fielded Park Geun-Hye, the former dictator's daughter, as their candidate.Amid worsening social inequality, this "moderniser" versus "democratiser" clash was dividing the country along the lines of political generations rather than socio-economic status (Joo 2017, 374-379).

Legacy of Employer Co-ordination
Hong Kong has exemplified LME characteristics since the 1950s, when the modern phase of industrialisation ensued.Neither the British colonial government nor larger corporations made serious efforts to facilitate market co-ordination.The colonial government first advocated a laissez-faire economic ideology, and later, one of "positive non-interventionism" (Scott 1986, 455-459).It did not use financing to steer the economy and the handful of institutional supports it did provide (such as the Export Credit Insurance Corporation and Productivity Council) were introduced in the late 1960s when export-led industrialisation was already flourishing.In labour matters, it relied on the archaic Employers and Servants Ordinance, enacted in 1902, until changing social circumstances forced it to introduce the Employment Ordinance in 1968 (England and Rear 1981, 196-246).
Business enterprises that operated in 1950s-1990s Hong Kong can be divided into three main groups: British hongs that dominated the finance, trade, and business services sectors 11 ; Shanghai emigr e industrialists who left the mainland as the communists gained control and introduced large-scale textile spinning and weaving establishments to Hong Kong; and numerous SMEs that were founded in response to manufacture-for-export opportunities conveyed by trading firms from Britain and Japan.There was no long-term co-operation on skills development or exchange of financial support among firms of different sizes (see Sit and Wong 1989;Chu 1992).Although some British hongs recognised the enterprise unions' rights of representation, neither the British hongs nor large Chinese enterprises provided substantial shop-level training (England and Rear 1981, 69-94).
Globalisation twice triggered deregulation in Hong Kong: first in 1979, when China was reintegrated into the capitalist world economy; and later during the 1997-1998 economic crisis.The former led to the massive dislocation of manufacturing workers, and the latter contributed to the casualisation of white-collar and professional employees.
In the absence of government leadership, industrial enterprises were unable to solve the collective action problem of technological investment and engage in industrial deepening.The introduction of the Employment Ordinance (1968) and Waste Disposal Ordinance (1980) also raised production costs.Given the dearth of firm-specific training for manufacturing workers and their low level of unionisation (11% in 1976), industrialists had no qualms about taking advantage of China's Open-Door Policy.Since 1979, Hong Kong has become the mainland's largest source of realised foreign direct investment, amounting to a cumulative value of US$1,099.2 billion or 54% of the total in 2018.Reflecting this change was a sharp decline in manufacturing employment from 1,025,009 persons in 1981 to 489,900 persons in 1996 and a decrease of manufacturing's contribution to GDP from 22.8% to 6.9% during the same period (HKCSD 1986;2001).
The hollowing out of Hong Kong's manufacturing industries did not lead to a surge in unemployment, although many manufacturing workers had to settle for unskilled service sector jobs.It also led to the shrinkage of opportunities at the middle level: in 1985, the salaries of technicians and supervisors in the wholesale/retail and personal services sector were lower than those of their counterparts in the manufacturing industry (HKCSD 1986, 40).Together with the small number of jobs created by the highly paid financial services sector (Lee, To, and Yu 2018), the territory experienced its steepest rise in the Gini coefficient during this period, soaring from 0.453 in 1986 to 0.518 in 1996 (HKFSO 2017).
The low level of market co-ordination persisted and a trend of neo-liberalisation set in after 1997.Although the newly inaugurated Special Administrative Region (SAR) government experimented with market co-ordination by pledging to promote technologyintensive industries, this initiative was not a success (Wang 2018, 407).Between 2005 and 2016, investment in research and development hovered at around 0.8% of GDP, and the contribution from private enterprises decreased from 51.5% to 43%, much lower than Korea's private sector contribution of 77% (Chu 2020, 345).Similarly, business enterprises showed limited interest in firm-specific training.A 2004 study found that employers supported no more than 16 hours of training, a figure smaller than that for the UK (Tam andChiu 2010, 2209).
Already weak, market co-ordination was enfeebled further by the 1997-1998 economic crisis, which plunged Hong Kong into a recession that continued well into the mid-2000s.The fear of a budget deficit led the government to restructure, streamline, and downsize the civil service (Goodstadt 2014, 36).This practice was copied by large enterprises, which affected not only manual workers and technicians but also white-collar workers and professionals.Lower-level jobs were outsourced and managerial/professional jobs that once offered de facto life-tenure became fixed-term contracts with indeterminate futures.A survey found that the average employment tenure decreased from 8.49 years in 1995 to 7.45 years in 2004, a figure lower than in all OECD countries except the USA (Chiu, So, and Tam 2008, 693).A recent survey found that the highest labour turnover rate was registered amongst the VLEs, suggesting the absence of an internal labour market where one might be expected (HKIHRM 2019).

Power Resources of Labour and Societal Corporatism
The colonial government's liberal-market orientation also guided its policies on organised labour.Despite its concern in the 1950s that unions affiliated with the Chinese Communist Party and Taiwan's Kuomintang would engage in open conflicts, it imposed few restrictions on the registration of trade unions at the enterprise, industrial and occupational levels (England and Rear 1981, 120-138).This, ironically, resulted in the proliferation of small unions crisscrossed with sub-ethnic and political identifications with little bargaining power.This largely explains the inability of workers to stem the two waves of deregulation examined above.
The level of unionisation had increased from around 13% in 1961 to about 20% in the late 1970s and, since the mid-1990s, 22-27%, a respectable level by international standards.However, the proportion of unions with 250 members or fewer increased from 52% in 1979 to 62% in 2018 (HKLD various years; Turner et al. 1980, 22-32).Furthermore, the economic sectors that provided the bulk of employment were the least unionised.Manufacturing, which hired some 45% of workers in most years between 1950s and 1980s, had slightly above 10% unionisation.Commercetrade, hotel and food service, finance, real estate, and business serviceswhich had predominated since the 1980s, also had below 10% union density for most years between 1976 and 2018 (HKLD various years).
A more acute source of weakness among Hong Kong's organised labour was its political divide.The pro-democratic Hong Kong Confederation of Trade Unions (HKCTU) emerged in 1990, seeking to build an autonomous union centre to represent workers' rights at the grassroots level (HKCTU 2019).It became an important challenger to the pro-Beijing Hong Kong Federation of Trade Unions (HKFTU).Although the HKFTU continued to command the affiliation of an overwhelming 37% to 47% of organised workers between 1991 and 2018, the HKCTU's 16% during the same period remained significant. 12 The political rift between the two deepened as Hong Kong approached the 1997 handover.Reflecting the HKFTU's commitment to implementing Beijing's policy of attaining a smooth handover of sovereignty and, after 1997, securing the SAR government's effective administration, it "made vocal commitments [in 1988] to the … need for cooperation and realism in the settlement of labour disputes with employers" (Chan and Snape 2000, 120).Whilst helping workers in claims, disputes, and accidents at work, and petitioning the government for more labour and welfare benefits, the HKFTU resisted efforts to empower organised labour.Not only did it remain aloof in HKCTU-aided labour protests against layoffs, low wages, and poor working conditions, but it also co-operated with pro-Beijing legislators in 1997 to suspend the HKCTU-initiated Ordinance, which gave labour unions the rights of representation, consultation, and collective bargaining (HKCTU 2019;ILO 1998).
The weakness and political division among organised labour were also evident in the working of the tripartite Labour Advisory Board (LAB), which the government established as a consultative body with business and labour representation.In part because of the method of electing labour representatives, none of the elected or appointed labour representatives since 1997 was HKCTU-affiliated (HKLD various years).Furthermore, the LAB was barely effective in facilitating employer-employee compromise, as demonstrated by the introduction of the minimum wage in 2012.
Specifically, the HKCTU voiced the need for a minimum wage in 1998, and, in 2001, public outrage was aroused by a news report on an old man working for HK$7 an hour (less than US$1) for a cleaning service outsourced by the government (Wong 2014, 737).By 2004, individual HKFTU leaders had also declared their support of a minimum wage.The government responded by commissioning the LAB for a consultative study of the matter.Yet after a year's deliberation, the LAB's employer representatives only agreed to introducing a charter that promoted the payment of decent wages among the business enterprises.In place of a charter, the government launched the 2006 Wage Protection Movement, the palpable failure of which, together with the emergence of overwhelming public support, forced the government to initiate a minimum wage bill for debate within the Legislative Council in 2008 (Wong 2014, 738).

Nature of the Political System
Hong Kong's political institutions and faultlines have, like Korea, been structured by geopolitics.The colonial government put off democratisation until the mid-1980s in anticipation of the 1997 handover.The Basic Law, introduced at handover as a quasi-constitution, promised the extension of electoral politics from District Board to Legislative Council and the Chief Executive; yet democratisation is predestined to be limited given Beijing's concern with ensuring the domination (or "effective administration") of the SAR government.By-products of Hong Kong's "electoral autocracy" are the disproportionate influence of capitalist interests and deflection of political effort from redressing socio-economic inequalities.
Constitutional design in a non-democratic setting is seldom a trivial matter (Gandhi and Przeworski 2006).Political institutions in authoritarian regimes are built to deliver policy concessions in exchange for co-operation.At the time of the drafting of the Basic Law, capitalist co-operation was imperative for the smooth transition of sovereignty.The Basic Law accordingly prescribes direct election for only 30 of the 60 Legislative Council seats.The remaining 30 seats are to be returned by functional constituencies, of which 21 represent business sectors or closely associated professions (such as accounting, construction).The 1,200-member Election Committee for the selection of the Chief Executive is similarly constituted.If the Basic Law solidified business' privileged position within the SAR, the mainland government adopted additional strategies of co-optation that facilitate business' direct access to Beijing and so augmented their influence.An example is membership of the National Committee of the Chinese People's Political Consultative Conference (see Fong 2014).Business tycoons have not hesitated to appeal directly to Beijing.For example, they petitioned against Chief Executive Tung Chee-hwa's ambitious public housing project that threatened to reduce the demands for private housing (Fong 2014, 213).All these moves have contributed to the neo-liberalisation of Hong Kong.
The extension of electoral politics has facilitated the emergence of pro-Beijing political organisations with strong labour connections, notably the HKFTU and the Democratic Alliance for the Betterment and Progress of Hong Kong (DAB).These organisations have become most adroit in electoral campaigns; together they captured 11 of the 35 geographic constituency seats in the 2016 Legislative Council election, with the DAB becoming the largest political party, with 12 of the 70 seats.Yet these effected no change in the Legislative Council's pro-capitalist structure.In identifying their key strategic interest as the pursuit of Beijing's Hong Kong policies, namely, "effective administration," the HKFTU and DAB refused to work with pro-labour parties in the democratic camp but chose to side with pro-Beijing business interests, and thus inadvertently acquiesced to neo-liberalisation.
The above shows how the HKFTU and DAB worked with other pro-Beijing legislators to roll back the labour unions' rights of representation and collective bargaining in 1997 (initiated by the pro-democratic HKCTU), and their slowness in supporting the minimum wage bill in the 2000s (also initiated by the HKCTU).In addition, both organisations succumbed to the neo-liberal logic and rendered staunch support for the Link Real Estate Investment Trust (REIT) initiative.In 2005, the fear of budget deficit led the government to sanction the privatisation of malls, wet markets, and parking lots in public housing complexes (the Link REIT deal).In addition to voting in favour of the government, the HKFTU and DAB mobilised protests against two pro-democratic legislators seeking judicial review of the government's decision (Chen and Pun 2007, 78;HK01, May 17, 2019).When the damage of the Link REIT deal to the livelihoods of public housing residents and small businesses became evident in 2008, most HKFTU and DAB legislators voted against the motion to "buy back the Link's shares" proposed by an independent legislator.Instead, they supported the motion to ask the government to issue a statement of concern (HKLC 2008).
Electoral politics requires the HKFTU and DAB to address labour and social needs.Their support for Leung Chun-Ying's bid to become the third Chief Executive, espousing a pro-poor agenda, was arguably an effort to defy business interests and facilitate farreaching redistribution.However, Leung's administration under-performed in public rental housing construction. 13In addition, Leung did not uphold his public pledge to introduce a "universal retirement benefit"; he only launched the "higher old age living allowance" with a means test at a much lower rate than promised in his Manifesto (Oriental Daily, September 22, 2011;Leung 2012).The DAB and HKFTU remained steadfast in their support of the government's "effective administration."Unwilling to work with parties in the democratic camp to initiate social dialogues for the remodelling of economic institutions in pursuit of more fundamental distributive measures, they had to be satisfied with stopgap measures as a means to maintain social stability.

Conclusion
Why are advanced capitalist societies in East Asia plagued by economic and social disparities even more severe than those found in some of their Western counterparts?Why do the manifestations of such disparities differ among high-income East Asian economies?This study has focused on Korea and Hong Kong, as exemplars of two pathways to highincome economic status, and argues that their severe social and economic disparities are best explained in terms of how their varieties of capitalism have unravelled under the pressure of globalisation.It examined the phenomenon with a framework that critically integrates the original VoC approach, Thelen's (2014) modification, and geo-political dynamics.
The findings on these two cases have been summarised in Table 1, but some of the major points can be highlighted here.In Korea, globalisation exacerbated dualisation partly because its state-imposed employer co-ordination nurtured weak employer-labour relations.In retaining high-quality manufacturing and maintaining firm-level coordination with their core workforce, much like their German and Japanese counterparts, Korean BBGs were also inclined to let the brunt of neo-liberalism fall on the peripheral firms and workers.In Hong Kong, globalisation unleashed sweeping deregulation because of the historical absence of employer co-ordination, which rendered expedient the wholesale relocation of manufacturing to China.
Korea's deepening dualisation can also be attributed to the presence of a small but powerful unionised labour force that, due to its political rift with the liberal left and defensive response to the BBG's neo-liberal practices, was unable to reach a policy compromise that could shield the non-core workers from harsh exploitation.Although superficially similar to the German case studied by Thelen (2014) in that unionised labour in the absence of social partnership presents the main obstacle to the ideal of "flexicurity," this study also contends for the need to examine the strategies of the business enterprises or to extend Thelen's political-institutional analysis beyond the domain of organised labour.By comparison, the explanation for Hong Kong's neo-liberalisation in extremis, even by the standards of the USA and the UK, is relatively straight-forward.Its weak and politically divided labour unions were unable to stem both the de-industrialisation and, later, the casualisation of white-collar/service employment.
This article also drew attention to Korea's and Hong Kong's pathways of democratisation as further explanations of their deepening inequality.Geo-politics set Korea on a path of conservative democratisation.In addition to preserving the domination of the BBGs and the pre-existing political organisations to the detriment of the emerging labour power, the mainstream liberal or left-of-centre forces also fell short of addressing essential socio-economic issues, but instead competed for voter support by focusing on Cold Warderived issues of national division and authoritarian legacy.The weakness of left-wing parties and labour unions was the political-institutional reason behind the absence of the flexicurity agenda in Korea.Nonetheless, Korea's liberal democracy has allowed its people to express their yearnings for redistribution and improvements in economic governance through the election of a liberal government in 2017 on the platform of redistribution ("income-led growth") and expanded social consultation. 14 In a similar way, geo-political considerations, especially China's interest in maintaining political stability and "effective administration" in post-retrocession Hong Kong, led the country to ratify an "unholy alliance" between pro-Beijing labour and business forces at the expense of labour interests.In non-democratic Hong Kong, even as political setbacks among the pro-Beijing groups and the need to deflect popular demands for liberal democratisation have sometimes been expressed as petitions for social welfare expansion, the entrenchment of neo-liberal values tends to emasculate such initiatives.
Apart from the empirical findings, this article also likes to make three theoretical observations.First, the original VoC approach has under-theorised the role of the state in employer co-ordination and, as a corollary, has not explored its implications for the CME's unravelling under globalisation (see also Kang 2010).The Korean case has demonstrated the weakness of employer-labour relations and their vulnerability in the face of neo-liberalism.Although the importance of the state is sometimes attributed to the uniqueness of East Asia, especially the region's geo-political positions and lateness in development, it is arguable that the state plays a more important role in many other cases.As such, future research might find it useful to theoretically explore the state-orchestrated employer co-ordination as an important variant of the CME.Second, although this article has appreciated the theoretical imperatives of Thelen's (2014) approach, it has nonetheless pointed to the need for the analysis to go beyond labour unions and political institutions within national boundaries.Just as geo-politics has exerted enduring influence over the central political debates and solidarity of organised labour in a society, business practices and strategies have also affected the willingness of workers to compromise.
The third point concerns the acuteness of dualisation and deregulation in Korea and Hong Kong, which contrasts sharply with Iversen and Soskice's (2019) postulation on the symbiosis between advanced capitalism and advanced democracy.Indeed, even as Korea has undergone the consolidation of democracy, the political accountability of the BBGs is weak.This is not to mention the imposition of electoral autocracy in Hong Kong.The vested interests in both cases had not stifled economic dynamism as in the "middleincome trap" societies, yet geo-political forces limited the extent to which the politics of democratisation could force the business interests to share their profit.Equally important, despite Iversen and Soskice's argument concerning the need for capital and skilled labour to co-locate in an advanced, knowledge-based economy, Korea and Hong Kong have built extensive production networks in East and Southeast Asia.This suggests the inadequacy of an argument founded overwhelmingly on the technological "imperatives" of production.Apart from the relative employer-labour power resources, this study also points to the importance of geo-political considerations.Not only had the Korean state supported the retention of core manufacturing capacity at home, a discussion on the need for reshoring also unfolded in recent months in light of USA-China tensions.These concerns have never surfaced in Hong Kong.This article hence suggests the need to situate an examination of the relationships between advanced capitalism and advanced democracy within the appropriate historical and geo-political contexts.

Disclosure Statement
No potential conflict of interest was reported by the authors.Notes 8.For example, representatives of labour unions and employers on average constituted about 30% of the tripartite committee membership while government officials and "public interest" representatives constituted about 70% (Kim and Ahn 2018, 16).9.The eight key tasks were: (i) win-win inter-generational employment; (ii) win-win employment between primary contractors and sub-contractors; (iii) win-win between regular and non-regular workers; (iv) introduction of a performance-based wage system; (v) reduce working hours; (vi) clarification of employment contracts; (vii) expansion of the social safety net; and (viii) strengthening social partnership to establish "productive bargaining culture" (KMOEL 2015).10.Mobrand (2018a, 619) notes that future president Moon Jae-In was one of four opposition liberal legislators who abstained.11.Hong is the Cantonese pronunciation of the Chinese character for a commercial firm.12.The HKCTU was dissolved on October 3, 2021.13.An annual average of 12,530 units were built under Leung (2012Leung ( -2017)), as compared with 14,034 units during his predecessor's tenure (HKHA 2022;HKLCS 2013;HKRVD 2019).14.To widen representation, President Moon Jae-In expanded the social dialogue to include "outsider groups" (small business, youth, women, and non-regular labour).However, despite electoral successes (2017,2018), the key reforms of the "labour respecting agenda" (minimum wage, working hours reduction, and public sector decasualisation) were seriously diluted under business pressure.
The social dialogue lost momentum (the KCTU did not join at all).Competent Covid-19 pandemic management led to an impressive victory for the governing party in the 2020 national assembly election but the government's support waned thereafter.Key to this was the spiralling property prices that alienated younger voters in particular.Reminiscent of 2007, disappointment with the liberal government's social performance paved the way for a populist conservative presidential candidate promising to revive growth by pro-market measures in 2022.

Figure 2 .
Figure 2. GDP per capita and Gini coefficient Source: World Bank database for GDP per capita (in constant 2010 US$) and GINI index; JILPT (various years); and HKCSD (2012).

Table 1 .
Pathways of globalisation: modified VoC explanation and East Asia Note:a.Indicators of dualisation: (i) regular versus non-regular composition of the labour force; (ii) wage disparities between regular and non-regular employees; (iii) LE and SME; and (iv) youth unemployment.