Multifaceted effects of globalisation on welfare attitudes: When winners and losers join forces

Abstract

How does economic globalisation influence individuals’ welfare state preferences? Moving beyond the unidimensional understanding of globalisation exposure, we intersect two dimensions of exposure perceptions (gain vs loss and individual vs societal impacts) and propose a novel typology: collective winner, lone winner, lone loser and collective loser. We then explain the preference gap among losers (collective losers vis-à-vis lone losers) and among winners (collective winners vis-à-vis lone winners) by considering three distinct motivations for welfare state support: compensation, risk-pooling and inequality reduction. We illustrate the usefulness of our typology using an original survey in South Korea. We find that lone winners are far more supportive of welfare spending than collective winners. At the same time, collective losers are found to be much more supportive of welfare spending than lone losers. We provide some first-cut evidence that the insurance-seeking motivation common to lone winners and collective losers drive their welfare state support.

Keywords

collective loser economic globalisation insurance seeking lone winner sociotropic perception South Korea welfare state attitude

Introduction

How does individuals’ exposure to economic globalisation influence their attitudes towards the welfare state? One major theoretical argument, often referred to as the compensation thesis, posits that citizens who suffer from globalised economic competition seek government protection. Depending on the size and political leverage of these ‘losers of globalisation’, the micro-level shifts in attitude might aggregate into an effective upward pressure for social spending and counter the race-to-the-bottom pressure allegedly exerted by market forces (Garrett, 1995; Rodrik, 1997, 1998). The dynamics of compensation seeking and granting are integral to ‘embedded liberalism’ in Western industrialised polities that have sought to use government social policies to buffer citizens from the vagaries of international economic openness and sustain popular support for an open economic order (Ruggie, 1982).

A number of empirical studies have tested the compensation thesis by examining public opinion in and across Western Europe and North America (Cusack et al., 2006; Margalit, 2011; Scheve and Slaughter, 2004; Walter, 2010), as well as in developing economies (Koster, 2014), and found, at best, mixed empirical support for the thesis. To reconcile the mixed or weak findings, scholars have proposed more nuanced versions of the thesis by differentiating distinct aspects of globalisation and/or distinct functions of the welfare state (Burgoon, 2001; Busemeyer and Garritzmann, 2017; Hays et al., 2005; Kim, 2007; Walter, 2017), by factoring in various contextual conditions that affect the political representation and economic conditions of losers (Burgoon, 2012; Ha, 2008; Rudra, 2002) and by taking into account existing means of protection within and outside the welfare state (Cao et al., 2007; Hwang and Lee, 2014; Lim and Burgoon, 2020; Rickard, 2012; Schaffer and Spilker, 2016).

Despite the recent advancement in the literature, in our opinion, there are two important limitations to the existing scholarship. First, the focus in the empirical literature has been overwhelmingly on identifying losers under globalisation and their demand for compensatory and protective policies. The literature thus fails to capture the potentially multidimensional triggers of welfare support under globalisation, in particular by winners. Second and relatedly, previous studies have overlooked the potential mismatch between collective-level and individual-level exposure perceptions, which we believe is key to moving beyond the narrow perspective that pits losers against winners. For example, an individual may perceive that economic globalisation has a rather positive impact on oneself but not on others in society. Individuals with such mixed perceptions about the individual and societal impact of globalisation may have distinctive welfare attitudes from those who believe globalisation hurts the majority of the society including themselves.

To address the limitations, we intersect two dimensions of globalisation exposure perception – (1) gain versus loss perception and (2) self-centred versus collective-level perception – and propose four conceptual types: collective winner, lone winner, collective loser, lone loser (see Table 1). Collective winners are those who perceive that globalisation benefits (or does no harm to) themselves and many others in society; lone winners are those who perceive that globalisation benefits (or does no harm to) themselves but hurts many others; lone losers are those who perceive that globalisation hurts themselves but not most others; and collective losers are those who perceive that globalisation hurts both themselves and others.¹

Table 1.

Two-dimensional globalisation perception.

Effect on the nation	Effect on myself
Effect on the nation	Not bad	Bad
Not bad	Collective winners	Lone losers
Bad	Lone winners	Collective losers

The overarching aim of this article is to offer a multidimensional typology that provides a better understanding of individual attitudes towards the welfare state under globalisation, compared to the conventional unidimensional continuum of winners and losers. As we elaborate on in the theoretical framework section, some non-losers have reasons to seek welfare state expansion (lone winners), and some losers might be reluctant to support welfare spending (lone losers). Ultimately, this nuanced understanding of the globalisation and welfare preference nexus should help both scholars and policymakers to move beyond compensation politics that may unnecessarily pit the losers of globalisation against the rest.²

We empirically illustrate the usefulness of our typology using an original survey and survey experiments in South Korea. Few micro-level studies of the nexus have been conducted in non-Western settings often characterised by smaller and less mature welfare state institutions. The South Korean case thus helps broaden both the theoretical discussion about and empirical scope of the nexus between globalisation exposure and welfare attitudes (see the ‘Case selection: South Korea’ section for details on case justification).

Our data reveal that the four types of citizens (as classified in Table 1) in South Korea indeed exhibit distinctive welfare state preferences. We also find that their welfare attitudes are not consistent with what the conventional compensation-seeking mechanism would expect. Overall, once we break down the conventional winner–loser continuum, we find the losers were no more supportive of welfare state expansion than the winners. In more substantive terms, we find that lone winners are the most supportive of welfare state expansion among the four types, while lone losers are the least supportive. Although the distribution of the types should differ across countries depending on macro conditions, we believe the general framework we developed can be used in any given society to explore the globalisation-welfare attitudes nexus. This article provides theoretical foundations for comparative research in the future.

Theoretical framework

Previous studies have considered three primary mechanisms for welfare state support – compensation seeking, insurance seeking and inequality reduction. We argue that globalisation contexts faced by individuals trigger a distinct combination of the mechanisms. By focusing on the distinct mechanisms, we seek to explain the preference gap among losers (collective losers vs lone losers) and the gap among winners (collective winners vs lone winners). We by no means disregard the difference between winners and losers, which has been discussed extensively in the existing literature. Rather, this article aims to enrich our understanding of the globalisation and welfare preference nexus by systematically considering additional, within-group differences in attitudes.

The compensation-seeking mechanism suggests that being losers of economic globalisation is associated with higher support for the welfare state as those losers are expected to demand more government compensation.³ The insurance-seeking mechanism, however, focuses on the risks and uncertainties, not necessarily on the already incurred or imminent losses, from economic globalisation. The sense of increasing (or increasingly unknown) risks under a liberalised economy might generate future-oriented demand for public risk-pooling as an attempt to prevent harmful consequences ex ante (Busemeyer and Garritzmann, 2017).

Finally, the equality-seeking mechanism posits that globalisation shapes welfare attitudes via its effect on inequality perception (Lü and Scheve, 2016; Norton and Ariely, 2011; Scheve and Slaughter, 2001). Studies find that individuals generally prefer a less unequal society than the one they are currently living in, although how much causal impact such a preference has on their support for redistributive public policies varies across societies and over time (Lü et al., 2012; Lü and Scheve, 2016; Norton and Ariely, 2011). One situational factor that is empirically found to strengthen the inequality-reduction motive is increasing macro-level inequality in society (Dimick et al., 2017; Kerr, 2014; Schmidt-Catran, 2016). If people perceive that globalisation increases the level of inequality, they may support the welfare state out of the inequality-reduction motive.

We suggest that the three mechanisms for welfare state support have different levels of playing field – the collective or individual level – which allows us to theorise how multidimensional globalisation perceptions trigger each mechanism. First, the compensation mechanism primarily relies on individual cost–benefit analyses, and the society-wide impact of globalisation alone is unlikely to trigger the mechanism. In contrast, the equality-seeking mechanism focuses on the interaction between the individual and social impact of globalisation and is likely to be triggered by a perceived divergence between individual- and collective-level globalisation experiences: those who perceive that only a small segment of the society suffers/gains from globalisation want to reduce the globalisation-driven inequality and thus, support welfare state expansion.

Finally, the insurance-seeking mechanism also involves both levels but operates differently at each level; therefore, we have two sub-mechanisms within the insurance-seeking mechanism. First, the assessment of one’s own economic situation should motivate one to seek insurance against future uncertainties, regardless of social environments. In addition, the insurance-seeking mechanism also works at the sociotropic level. Individuals often infer their own economic prospects from their co-nationals’ situation,⁴ and even if an individual is faring well for now, she or he can take a cue from the dismal circumstances of many others under economic globalisation and recognise that the vagaries of an open economy may negatively affect her or him in the future. Thus, the perception of a society-level threat is more likely to provoke insurance needs than the lack of such a threat perception. Furthermore, social cues can signal a broad-based interest in public risk-pooling. Compared to those who think globalisation is innocuous for most citizens, citizens who perceive globalisation as a common threat are more likely to expect that other citizens will be on board with public risk-pooling. This perception of shared needs is vital for provoking insurance seeking via the welfare state as opposed to private saving.

In discussing the three mechanisms, it is important to note that we do not equate them with any specific underlying rational-materialistic or non-materialistic motivations. For instance, as for the equality-seeking mechanism, one might prefer a welfare policy reducing inequality out of his or her jealousy towards the better-off, compassion towards those who are even worse off, normative commitment to a more egalitarian society, rational aversion to unrest and disorder associated rising inequality (Rickard, 2015) or any combination of the aforementioned.

Expectation

We now hypothesise how variegated mechanisms for welfare state support – compensation, insurance and inequality reduction – play out differently within each set of the two groups: globalisation losers and globalisation winners.

Lone losers versus collective losers

Lone losers are those who perceive that economic globalisation only hurts themselves, not others in society, while collective losers are those who perceive that economic globalisation has a negative impact on themselves as well as others in society.⁵ Because both lone and collective losers perceive that they incur losses from economic globalisation, as existing studies drawing on the compensation thesis hypothesise, these citizens would seek compensation for losses they incurred in the open market.

Compensation seeking may not be the only mechanism that is observable among the losers. Because they think that globalisation hurts them, both types of losers would exhibit individual-level insurance seeking (e.g. via personal savings). Yet, collective losers, who see that others also suffer from globalisation, would additionally develop a preference for societal-level insurance seeking. The social cues about the losses faced by many co-nationals who are likely to be on board with societal risk-pooling trigger demand for public social programmes on top of private insurance seeking. To the contrary, lone losers who perceive that the rest of the citizens are faring well are not subject to the social cue. The societal-level insurance seeking is thus less likely to shape lone losers’ welfare support.

Instead, lone losers who perceive that they themselves are the suffering minority under globalisation may seek inequality reduction. The perception of uneven distributional consequences of globalisation within society may provoke feelings of distributive injustice or jealousy towards others who benefit or are sheltered from globalisation, leading lone losers to seek a remedy against perceived macro-inequality and demand redistributive welfare policies. In contrast, collective losers believe that economic globalisation causes collective suffering of the society as a whole, which means the perceived distributional implication of globalisation is not necessarily unequal. They are thus less likely to demand redistributive welfare policies than are lone losers.

From the discussion, we see that the main difference between lone and collective losers lies with society-level insurance-seeking and equality-seeking mechanisms, and we thus hypothesise the following:

H1: Lone losers are more likely to support a welfare state that reduces social inequality than are collective losers.

H2: Collective losers are more likely to support a welfare state for social risk-pooling than are lone losers.

Lone winners versus collective winners

How about those who think they benefit from economic globalisation? Since both lone and collective winners perceive that they themselves do well under globalisation, the compensation mechanism is unlikely to be applicable to winners in general. Still, although both types of winners think they fare well currently, it does not prevent them from seeking insurance from future uncertainties. In particular, lone winners perceive that many others in society suffer from the vagaries of globalisation. This perception is likely to trigger the insurance-seeking mechanism. Even if they benefit from the system for now, they can infer from others’ dismal circumstances that ‘tomorrow, it will be me’. The social cues render them more inclined to participate in risk-pooling schemes or subsidise other citizens (possibly including their future selves) who face risks. Compared to lone winners, collective winners who do not receive such social cues would have less incentive to participate in social risk-pooling.

Lone winners also perceive that globalisation’s distributional consequences are unequal within a country. Globalisation has been fine for them but left others worse off. Such a perception of increased macro-inequality may foster advantageous inequality aversion or altruism among lone winners and make them more supportive of redistributive welfare state expansion. Of course, lone winners need not be altruistic to support a redistributive welfare state because mitigating (perceived) inequality induced by globalisation could serve the lone winners’ material and strategic interests as well. For example, by agreeing to redistribute their disproportionate gains from trade, lone winners who have a vested interest in the open trade system can pre-empt the mobilisation of disadvantaged losers against the system (Rickard, 2015) and safeguard their personal property against potential disorder (David et al., 2016). This logic is also consistent with embedded liberalism, suggesting that redistributive institutions are integral to generating social acceptance of an open economic order (Ruggie, 1982). Thus, regardless of the underlying logic, lone winners who perceive that globalisation induces inequality are expected to be more supportive of the redistributive welfare state than are collective winners.

In sum, we posit that, all else equal, lone winners would be more supportive of the welfare state than are collective winners, because both social insurance-seeking and inequality-reduction mechanisms are more likely to be triggered among lone winners.

H3: Lone winners are more likely to support a welfare state that reduces social inequality than are collective winners.

H4: Lone winners are more likely to support a welfare state for social risk-pooling than are collective winners.

Table 2 summarises our framework. In essence, we suggest a comparison should be made not just between winners and losers but also within each group. For instance, our framework suggests that welfare state preferences would vary among the winners of globalisation. This approach should provide us with a better understanding of the distinct coalitions for/against welfare state expansion in different societies, compared to a simple dichotomy of losers versus winners.

Table 2.

Linking globalisation exposure perceptions to motivations.

	Motivations
	Individual-levelcompensation	Individual-levelinsurance	Society-levelinsurance	Society-levelinequality reduction
Lone loser	✓	✓		✓
Collective loser	✓	✓	✓
Lone winner			✓	✓
Collective winner

We do not have a prior expectation about which type of citizen is more prevalent within a society. But the framework of intersecting individual-level and collective-level globalisation perceptions should be general and flexible enough that it can be used in any given society to explore the variegated nexus between globalisation and welfare state attitudes. Future research can also explore systematic spatial (across societies) and temporal variation in globalisation perceptions and their association with different welfare-seeking mechanisms.⁶

Research design

To test the hypotheses, we conducted our original survey in South Korea in December 2016. Our empirical analyses have two goals. First, we show descriptive information about the distribution of the four types of citizens – collective winners, lone winners, collective losers, lone losers – in a society. As we highlight below, a substantial minority are loners who exhibit a mismatch between self-centred and sociotropic globalisation perceptions. Second, our survey embedding an experimental component is designed in particular to distinguish between lone and collective losers (i.e. within-loser variation) and between lone and collective winners (i.e. within-winner variation) in light of our theoretical expectations about the distinctive mechanisms driving their welfare support.

Case selection: South Korea

Studying South Korea has important empirical and theoretical merits. Empirically, there is still a dearth of micro-level studies exploring the globalisation and welfare state nexus outside of the West and especially in the newly industrialised/emerging economies in Asia. South Korea is as economically open as, if not more open than, many Western economies on which the compensation thesis has been tested. As of 2017, the Korean economy featured both higher import and export dependencies (36% and 41% of GDP, respectively) than the OECD averages (both 29%).

Yet, South Korea’s immature welfare state is distinguished from Western and in particular European welfare states, which might yield theoretically meaningful differences in how globalisation shapes welfare attitudes. We know from the welfare state scholarship that citizens’ experiences of having lived in one or the other type of welfare regime (Esping-Andersen, 1990; Hall and Soskice, 2001) lead them to form certain expectations of the government role in social protection. These shared expectations among citizens, in turn, explain why welfare attitudes vary between countries (Gingrich and Ansell, 2012; Jæger, 2009; Larsen, 2007), but see also Aarøe and Petersen (2014).

South Korea is a unique case in this regard because it is difficult to pinpoint what shared experiences that Korean citizens have with their welfare state. South Korea’s welfare state has been known as the ‘minimalist’/‘productivist’ welfare state, where welfare social protection was largely subordinated to economic growth (Haggard and Kaufman, 2008; Rudra, 2007) and ‘institutionalised social safety nets, if any, cover only a small segment of the productive workforce’ (Lim and Burgoon, 2020). Yet, the country’s welfare state is now at a crossroads. While some note the continuation of a minimalist/parochial welfare state (Yang, 2013), others focusing on the expansion of general tax-funded social services (An and Peng, 2016; Estévez-Abe and Kim, 2014) suggest that the country has moved closer to the social democratic welfare states of the Nordic countries (Kuhnle, 2004). It is thus reasonable to assume that Korean citizens do not share specific expectations about what the welfare state’s primary role is and should be. One commonality among South Koreans is that they have not yet experienced a highly institutionalised form of governmental social protection (Lim and Tanaka, 2020).

We thus contend that the Korean setting can provide a new and more neutral testing ground for our theoretical expectations. Furthermore, we believe that our findings from South Korea might be generalisable to other emerging welfare states in Asia that have increasingly opened up their economy. Ultimately, the extent to which the perceptions of on-going globalisation triggers demand for government intervention should vary depending on the different experience of welfare states. In this sense, applying our analytical framework to a broader set of countries can be a valuable addition for future comparative research.

Survey design and operationalisation of variables

We designed an original survey and conducted it online with a sample of 1804 Korean adults in December 2016. Our sample was drawn by a Korean survey firm, opensurvey, from their opt-in online panel.⁷ Stratified sampling was employed to approximate the parameters of the Korean population in terms of gender, age and residential locations at the metropolitan city/provincial level, but not in terms of other socio-economic indicators such as education and income. To alleviate concerns of sample selection bias with the non-probability sample, we also provide descriptive statistics of our sample compared to the Korean population in 2016 (based on government statistics) in the Supplementary Information.⁸

Independent variable: Globalisation attitudes

Our survey includes two questions on the perceived consequences of economic globalisation. One question is concerned with globalisation’s consequences on their own economic insecurity: ‘What do you think of the effect of economic globalisation (free trade, investments, and movement of workers) on your own economic situation?’ Responses are given on a five-point scale: 1 = a very good effect, 2 = a good effect, 3 = the effect is neither good nor bad, 4 = a bad effect to 5 = a very bad effect. To examine the impact of globalisation on the nation rather than on individuals, a separate question asks about the consequences of globalisation on the employment insecurity of Koreans as a whole: ‘Economic globalisation (free trade, investments, and movement of workers) threatens Koreans’ jobs’. The respondents answer on a five-point scale: 1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree.

To create four nominal types of globalisation perception based on the two 5-category ordinal indicators, we treat the neutral response (three in the ordinal scale) as well as the positive responses (one or two in the ordinal scale) as globalisation optimism in a relative sense. Using the operationalisation scheme, we code the four types of citizens in the following ways. The collective winners are those exhibiting optimism at both individual and collective levels. The lone winners hold an optimistic individual-level perception, but a pessimistic collective-level perception. The lone losers are the opposite: they have an optimistic collective-level perception yet a pessimistic individual-level perception. Finally, collective losers exhibit negative individual- and collective-level perceptions. We may also use different coding schemes; for example, treating the neutrals separately from the winners results in seven ideal types. To minimise concerns that our results may be sensitive to the differences in the coding, we report the results from employing this alternative typology in Table A.1 in the Supplementary Information. The alternative typology does not improve the model fit while our key findings do not change substantially (see Table A.2 and Figure A.1 in the Supplementary Information).

Dependent variable I: Support for social spending

Our survey asks whether respondents support or oppose a hypothetical income transfer proposal. Responses are given on a five-point scale ranging from 1 = strongly oppose to 5 = strongly support.⁹ One caveat is that when asked about support for a social programme, some respondents may think it does not have fiscal implications while others readily incorporate fiscal burden into their answers. To address the issue, we explicitly emphasise the fiscal burden aspect of social spending and take a two-step approach in collecting the respondents’ level of support.¹⁰ In this way, we can control for the unobservable individual-level heterogeneity in fiscal awareness and fiscal illusion.

More specifically, we first present respondents with a hypothetical scenario that explicitly externalises costs: ‘This is a hypothetical scenario. Your local government has a significant budgetary surplus this year from successfully hosting an event.¹¹ The government has proposed to spend the surplus money on an income transfer programme’.¹² We then ask, ‘What do you think of the proposal?’ After the first step, we introduce the fiscal constraint to the scenario and ask their preference once again: ‘The budgetary surplus for this year is rather unusual. If the government planned to continue this transfer programme regardless of the budgetary situation, what would you think of the plan?’ We use the responses to this second-step question as our primary dependent variable and call it support for social spending. To examine the mechanisms for the support, we also use the same dependent variable but with an experimental setting (see below) and another variable called support robustness (see also below).

Dependent variable II: Support for targeted and universalistic transfers

We suggested that two mechanisms for welfare support are the key to understanding the within-winner (within-loser) variation in welfare attitudes: one that prioritises inequality reduction and one that primarily focuses on social risk-pooling. To empirically adjudicate between the two mechanisms, we randomly assign respondents to one of the three hypothetical spending proposals: (I) universalistic spending in which all residents receive a flat-rate transfer; (II) moderately means-tested spending in which those earning less than 80% of the median income are eligible for the transfer; and (III) narrowly means-tested spending in which only those earning less than 40% of the median income are eligible.¹³

From our theoretical argument, we expect a stronger inequality-reduction preference to be associated with a more positive stance towards the pro-poor targeted transfer.¹⁴ An observable implication is that an equality-seeking subset of our sample would be more supportive of pro-poor targeting than the rest of the citizens. More specifically, support for social spending under treatments II and III would be higher among lone losers than among collective losers (H1) and among lone winners than among collective winners (H3). By contrast, those who are interested in social risk-pooling have a reason to support a universalistic transfer. Such a programme serves as a preventive social safety net for the insecure who are not necessarily disadvantaged (Rehm et al., 2012). If so, support for social spending under treatment I should be higher among collective losers than among lone losers (H2) and among lone winners than among collective winners (H4).

Dependent variable III: Acceptance of fiscal burden

We also test an additional observable implication of our argument with regard to fiscal burden, which should increase confidence in our empirical tests. It is only natural to expect that citizens driven by distinct mechanisms for welfare state support vary in their acceptance of the fiscal contribution to the welfare state. For instance, stronger demand for the social risk-pooling function of the welfare state would be associated with a greater acceptance of the need for fiscal contribution. Risk-pooling by definition requires participants’ contributions. Also, those citizens who wish to mitigate their advantageous inequality would be more accepting of fiscal contribution. It is their increased contribution to the redistributive welfare state that can most effectively realise the desired outcome of lower inequality.

To capture one’s attitudes towards the welfare state’s fiscal burden, we create the support robustness variable. The variable is constructed by taking the difference between support for social spending and the initial support for the income transfer under the budget surplus. It ranges from −4 (when a strong supporter under the budget surplus becomes a strong opponent) to 4 (when a strong opponent under the budget surplus becomes a strong supporter). A negative value of support robustness captures the fiscal burden-induced decrease in support. For instance, if a respondent who strongly supported (= 5) social spending under the budget surplus changes her answer to the neutral (= 3) option under fiscal constraint, then the value of support robustness for this respondent is 3 − 5 = −2.

In light of our hypotheses, we expect collective losers with a social risk-pooling need to exhibit stronger support robustness than lone losers (H2). Also, compared to collective winners, lone winners with advantageous inequality reduction (H3) and/or social risk-pooling (H4) needs should show stronger support robustness. Table 3 summarises our empirical expectations.

Table 3.

Empirical implications.

	Implications
	Support-level targeted transfer	Support-level universal transfer	Support robustness
Lone loservs.Collective loser	HigherLower	LowerHigher	LowerHigher
Lone winnervs.Collective winner	HigherLower	HigherLower	HigherLower

Controls

Our survey includes basic demographic questions on gender, age, income and level of education. We also control for income because, provided that individuals understand progressive taxation, being in the position of net tax contributors might dampen welfare support. Also included is an ordinal variable measuring individuals’ access to private protection in the model specifications: (1) no private pension insurance or pension savings subscription, (2) one subscription and (3) two or more. This variable is necessary because individuals may seek protection not by demanding welfare support from the government but by demanding from their employers better protection or by relying on market-based risk-pooling (Lim and Burgoon, 2020; Potrafke, 2019).

We also control for the five-point scale of the ideological self-placement variable: 1 = conservative to 5 = progressive. Finally, we control for respondents’ assessment of the local government’s administrative performance (ranging from 1 = doing very poorly to 5 = doing very well). Respondents may be reluctant to support a proposal they find desirable when the government is perceived to be lacking the capacity to administer it. Table 4 reports the descriptive statistics.

Table 4.

Descriptive statistics.

	Minimum	Median	Mean	Maximum	Obs.
Ordinal globalisation perception
Individual effect	1	3	3.14	5	1804
National effect	1	3	3.30	5	1804
Support for social spending
Pooled	1	3	3.06	5	1804
Universal flat rate	1	3	3.26	5	601
Moderate targeting	1	3	3.00	5	603
Narrow targeting	1	3	2.91	5	600
Support robustness	−4	0	−0.27	4	1804
Controls
Gender	1	2	1.51	2	1804
Age	20	43	42.47	69	1804
Education	1	4	3.91	5	1804
Income (deciles)	1	6	6.40	10	1804
Ideology	1	3	3.31	5	1804
Private pension	1	2	1.64	3	1804
Government performance	1	2	2.02	5	1804

Empirical analysis

Distribution of four types

Table 5 summarises the distribution of answers across the two globalisation perception questions in light of the typology we proposed. The table shows that, while collective winners are the largest group (47.6%) in South Korea, there is a sizable number of lone winners who perceive that globalisation hurts other co-nationals (21.4%). Losers also vary in their sociotropic perceptions. The lone losers (11.9%) who are typically associated with the compensation hypothesis are a minority group in South Korea and are outnumbered by collective losers (19.1%).

Table 5.

Globalisation perceptions in South Korea (sample distribution, n = 1804).

Threat to Koreans	Effect on me
Threat to Koreans	Not bad	Bad/very bad
Not a threat	Collective winner, 47.6%	Lone loser, 11.9%
Threat (agree/strongly agree)	Lone winner, 21.4%	Collective loser, 19.1%

Main findings: Distinct mechanisms triggering welfare support

Based on the discussion above, we now turn to examining what motivates each type’s welfare support. Given the ordinal nature of our dependent variables, we employ an ordered logit regression as our primary estimation strategy. Table 6 reports our main analysis. The key independent variable is globalisation perception. In all the models, we use a four-category nominal variable of which the baseline category is collective winner. Model 1 is our baseline model, employing support for social spending as the dependent variable. The coefficient estimate for lone winner indicates that lone winners are more likely to support welfare spending than collective winners, and the effect is statistically significant at the 1% level. We also find that collective losers are more supportive of social spending than collective winners, though the level of statistical significance is lower, at the 10% level.

Table 6.

Effect of globalisation perception on welfare attitudes.

	1	2	3
Globalisation perception categorical (baseline: collective winner)
Lone winner	0.368***	0.517***	0.199***
Lone winner	(0.111)	(0.193)	(0.053)
Lone loser	−0.195	−0.234	−0.021
Lone loser	(0.139)	(0.243)	(0.065)
Collective loser	0.222*	0.179	0.092*
Collective loser	(0.118)	(0.214)	(0.055)
Lone winner × targeted		−0.227
Lone winner × targeted		(0.235)
Lone loser × targeted		0.05
Lone loser × targeted		(0.297)
Collective loser × targeted		0.082
Collective loser × targeted		(0.256)
Controls
Initial support			−0.389***
Initial support			(0.021)
Low income targeted		−0.539***
Low income targeted		(0.133)
Gender	0.156*	0.165*	0.056
Gender	(0.088)	(0.089)	(0.042)
Age	0.018***	0.018***	0.007***
Age	(0.004)	(0.004)	(0.002)
Level of education	−0.022	−0.024	−0.027
Level of education	(0.066)	(0.067)	(0.031)
Income	−0.066***	−0.069***	−0.042***
Income	(0.024)	(0.024)	(0.011)
Ideology	0.152***	0.163***	0.018
Ideology	(0.043)	(0.043)	(0.020)
Private pension	−0.128**	−0.138**	−0.024
Private pension	(0.061)	(0.061)	(0.029)
Government performance	0.021	0.017	−0.028
Government performance	(0.050)	(0.050)	(0.023)
Constant			0.978***
Constant			(0.195)
N	1804	1804	1804
DV: Support for social spending	✓	✓
DV: Support robustness (to fiscal burden)			✓

p < .1; **p < .05; ***p < .01.

The predicted support levels based on Model 1 are visualised in Figure 1. All other covariates are set at the sample median values. In this and other graphs, we report 90% confidence intervals. The figure reveals that, in South Korea, lone winners have the highest support for social spending (43%), followed by collective losers (40%). Overall, these findings suggest that each type has a distinctive support level for the welfare state and justifies our call for moving beyond the loser-versus-winner approach in understanding the nexus between globalisation and welfare attitudes.

Figure 1.

Support for social spending (based on Model 1).

Now, we test our hypotheses about the type of welfare state supported by each group. Recall that all our hypotheses are stated in relative terms in comparison with other types of citizens. In Model 2, we employ a dummy variable targeted to indicate those respondents in the treatment groups who read about a pro-poor targeted transfer proposal.¹⁵ We then interact the dummy variable with the globalisation perception indicators. Again, following our discussion above, the intuition is that respondents whose globalisation exposure provokes an inequality-reduction mechanism would be more supportive of pro-poor targeting than other respondents. Also, respondents who develop social risk-pooling needs under globalisation would be more supportive of universalistic transfer than those without such a trigger.

Figure 2 visualises our finding, where the blue-coloured bars capture the support for the pro-poor targeted income support proposal. Among the losers, collective losers are more supportive of the targeted transfer than lone losers, which is not consistent with the observable implication of H1, which posits that lone losers are more likely to support a welfare state that reduces social inequality than are collective losers. Among the winners, lone winners are more supportive of the targeted transfer than are collective winners, which is in line with H3: lone winners are more likely to support a welfare state that reduces social inequality than are collective winners.

Figure 2.

Effect of pro-poor targeting on support for social spending.

When it comes to the universalistic flat-rate transfer (the red-coloured bars), collective losers are more supportive of the targeted transfer than lone losers, which is consistent with H2: collective losers are more likely to support a welfare state for social risk-pooling than are lone losers. Among the winners, lone winners are again more supportive than are collective winners. This result is consistent with H4: lone winners are more likely to support a welfare state for social risk-pooling than are collective winners.

In Model 3, we examine whether our expectations about support robustness hold. By construction, the value of this dependent variable can be more negative for those with a high level of initial support. We thus control for the initial level of support.¹⁶ Again, our intuition is that individuals motivated by advantageous inequality reduction and/or social risk-pooling would exhibit more robust support and be more accepting of fiscal burden than individuals without such motives. Within the group of winners, we thus expect lone winners to be more subject to both risk-pooling and advantageous inequality-reduction mechanisms and show greater support robustness than do collective winners (recall H3 and H4). Among the group of losers, support robustness would be higher for collective losers with stronger risk-pooling needs than lone losers without such needs (recall H2).

Our findings based on Model 3 are visualised in Figure 3. The y-axis values correspond to predicted support robustness, or in more substantive terms, the predicted change in support upon the introduction of fiscal burden. All other covariates are set at the sample median values. When individuals are told that the income transfer can no longer be financed with the budget surplus, collective winners reduce their support much more than do lone winners. In other words, lone winners’ support for the income transfer is more robust, which is consistent with H3 and H4. Comparing lone losers and collective losers, the support among collective losers is more robust. The finding is in line with H2.

Figure 3.

Support robustness.

Overall, all our hypotheses are supported by the empirical evidence except for H1. H1 posited that lone losers motivated to reduce their disadvantageous inequality should be more supportive of pro-poor targeted benefits than collective losers. Our evidence contradicts this expectation. Furthermore, Figures 1 and 2 show that lone losers, surprisingly, are the least likely to support any type of social spending. The predicted probabilities reported in Figure 2 suggest that only 37% of lone losers support a universalistic transfer, and only a quarter support a pro-poor targeted transfer. There is a nearly 10%-point gap in support between lone and collective losers who, in theory, should share similar preferences for compensation demand.

In the Supplementary Information, we explore several tentative explanations for the lone losers’ attitudes while leaving more rigorous testing for future research. One plausible explanation, also suggested by Colantone and Stanig (2018), is that some losers of globalisation may not find the promise of redistribution credible. Lone losers who find themselves to be a disadvantaged minority under globalisation may feel ostracised by the rest of the society, which would discourage them to demand compensation. In line with the expectation, our additional analyses reported in the Supplementary Information find that the lone losers generally exhibit a lower level of social trust than the rest of society, and it is these lone losers with low levels of trust who exhibit particularly sceptical attitudes towards the proposed income transfer programme.

It is also plausible that the lack of proactive demand for compensation from lone losers is driven by their economic conservatism. While the government protection of the disadvantaged might be a valence issue to a certain degree in mature welfare states, citizens in productive welfare states in Asia might still think the failure to adjust to the globalised economy is an individual problem that should be solved on their own, not through government spending (Lim and Burgoon, 2020). In the Supplementary Information, we present a model where we include the interaction term between ideology and globalisation perception. We indeed find that the lone loser’s distinctively low support for welfare state is most salient among conservatives. Still, across the ideological spectrum, collective losers and lone winners are more supportive of the welfare state than lone losers. See additional analyses on lone losers in the Supplementary Information for a more detailed discussion on these models and a visualisation of key findings.

Conclusion

Many scholars have studied the nexus between globalisation exposure and welfare demands. However, little is known about how individual-level and national-level consequences of globalisation interact to shape welfare attitudes among both winners and losers. This article proposed four types of citizens based on self-perceived economic consequences of globalisation: collective winner, lone winner, lone loser and collective loser. The typology allows us to identify important variation in welfare support within winners as well as losers.

Empirical analyses based on our original survey in South Korea find that there is indeed important variation in support for the welfare state among winners and also among losers. As for winners, lone winners are distinctively more supportive of the welfare state than collective winners. The highest level of support for the Korean welfare state comes from lone winners and, specifically, those lone winners treated with universalistic (as opposed to pro-poor targeted) social spending. Among losers, collective losers are distinctively more supportive of the welfare state than lone losers. Collective losers’ welfare state support is the second highest among the four types following lone winners. Both findings indicate that Korean citizens are more motivated by collective level concerns about globalisation. Among different mechanisms linking globalisation with welfare support, our empirical analyses lend the most consistent empirical support to the insurance-seeking mechanism.

We believe our findings in the context of South Korea carry important policy implications. Our study reveals that a sizable minority of citizens classified as lone winners have the potential for driving the expansion of the welfare state, especially the adoption of universal preventive policies. In South Korea, the size of lone winners (21% of respondents) was indeed twice the size of lone losers (11% of respondents) (recall Table 5).

If lone winners were to drive welfare state expansion under globalisation, what would the ramifications be for the protection of the poorest and disadvantaged? Preventive social policies preferred by lone losers might have limited progressive redistribution potential. At the same time, the literature suggests that a weaker overlap between the traditional supporters of redistribution (the disadvantaged) and the insurance seekers (the insecure) can cultivate broader-based support for the welfare state (Rehm et al., 2012). Once globalisation broadens the support base for the welfare state, the type of policy frames employed by political leaders and policy entrepreneurs could play a critical role in determining the direction of welfare state expansion (Lim and Tanaka, 2019; Béland and Cox, 2016). Given our findings, at least in South Korea, framing new social spending as future-oriented policy might be more effective in tapping into lone winners’ willingness to support welfare policies than framing the spending as pro-poor or inequality-reducing. The former type of framing could help to broaden the fiscal support base for social spending, which, in turn, allows the government to pursue compensatory or egalitarian goals via increasingly generous, universalistic benefits (Korpi and Palme, 1998).

To be sure, our specific empirical findings in the context of South Korea may not be fully generalisable to other settings. Yet, we still believe that South Korea is representative of the productivist welfare states in Asia and that our empirical findings can travel to the region’s emerging welfare states. Future studies should explore theoretically interesting and systematic between-society variations in the distribution of the four exposure types and in the relative strength of different mechanisms.

There can also be important temporal variation in the globalisation-welfare attitudes nexus. For example, Nooruddin and Simmons (2009) suggest that the level of uncertainty created by globalisation varies by the degree of openness. If so, the insurance-seeking motive as a driving force of welfare state expansion may give way to other mechanisms as a country’s globalisation deepens. Finally, future research can also investigate within-country variation in the typology. We hypothesised that the existence of collective-level threat influences welfare state attitudes. It is possible that this collective-level threat perception is driven by the regional-level exposure to globalisation shocks (Colantone and Stanig, 2018). If this is the case, the distribution of the four types of citizens would vary between struggling and sheltered regions of the same country, which leads to varying levels and types of welfare demand in different regions.

We believe future research that utilises our analytical framework would help map out the variation in the globalisation-welfare attitudes nexus. Understanding such variations would allow us to have a better picture of the distinct coalitions for/against welfare state expansion in different societies across different time periods.

Supplemental Material

sj-pdf-1-bpi-10.1177_1369148120974885 – Supplemental material for Multifaceted effects of globalisation on welfare attitudes: When winners and losers join forces

Supplemental material, sj-pdf-1-bpi-10.1177_1369148120974885 for Multifaceted effects of globalisation on welfare attitudes: When winners and losers join forces by Sijeong Lim and Seiki Tanaka in The British Journal of Politics and International Relations

This article is distributed under the terms of the Creative Commons Attribution 4.0 License (https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).

Footnotes

Acknowledgements

The authors thank Brian Burgoon, Jaemin Shim and participants of seminars at the East Asia Institute (EAI) for their valuable comments. They also thank the MacMillan Center of Yale University, EAI and the Center for Global Partnership of the Japan Foundation for generous support for this project. The human subject protocol of the research was evaluated by the Amsterdam Institute for Social Science Research (AISSR) Ethical Advisory Board.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This project was funded by a research grant from the Political Economy and Transnational Governance (PETGOV) Programme Group at the University of Amsterdam.

ORCID iD

Sijeong Lim

Supplementary information

Additional supplementary information may be found with the online version of this article.

Contents

Robustness Check with an Alternative Typology.

Table A.1: Alternative Globalization Perception Typology (N = 1804).

Figure A.1: Social Spending Preference based on Alternative Typology.

Table A.2: Replication with Alternative Globalization Perception Typology.

A Comparison between our Sample and the General Population.

Income.

Table A.3: Sample comparison for income.

Education.

Table A.4: Sample comparison for education.

Table A.5: Replication of Model 2 Using Weighted Data.

Figure A.2: Effect of Pro-poor Targeting on Support for Social Spending (Weighted Sample).

Additional Analyses on Lone Losers.

Figure A.3: Conditioning Effect of Trust.

Table A.6: Additional Models.

Figure A.4: Conditioning Effect of Ideology.

Figure A.5: Migration Attitudes (Descriptive Statistics).

Brant Test and Nominal Effects Model.

Table A.7: Brant Test for the Parallel Regression Assumption.

Table A.8: Model 2 with Nominal Effects.

What Determines Globalization Exposure Perceptions?

Table A.9: Determinants of Globalization Threat Perceptions.

References

Notes

References

Aarøe

Petersen

(2014) Crowding out culture: Scandinavians and Americans agree on social welfare in the face of deservingness cues. Journal of Politics 76(3): 684–697.

Peng

(2016) Diverging paths? A comparative look at childcare policies in Japan, South Korea and Taiwan. Social Policy & Administration 50(5): 540–558.

Béland

Cox

(2016) Ideas as coalition magnets: Coalition building, policy entrepreneurs, and power relations. Journal of European Public Policy 23(3): 428–445.

Bergh

(2019) The compensation hypothesis revisited and reversed. IFN working paper no. 1273. Available at: https://www.ifn.se/publikationer/working_papers/2019/1273 (accessed 21 November 2020).

Burgoon

(2001) Globalization and welfare compensation: Disentangling the ties that bind. International Organization 55(3): 509–551.

Burgoon

(2012) Partisan embedding of liberalism: How trade, investment, and immigration affect party support for the welfare state. Comparative Political Studies 45(5): 606–635.

Busemeyer

Garritzmann

(2017) The effect of economic globalization on compensatory and social investment policies compared: A multi-level analysis of OECD countries. DaWS Working Paper Series, No. 2, 29 November. Odense: University of Southern Denmark.

Cao

Prakash

Ward

(2007) Protecting jobs in the age of globalization: Examining the relative salience of social welfare and industrial subsidies in OECD countries. International Studies Quarterly 51(2): 301–327.

Colantone

Stanig

(2018) The trade origins of economic nationalism: Import competition and voting behavior in Western Europe. American Journal of Political Science 62(4): 936–953.

10.

Cusack

Iversen

Rehm

(2006) Risks at work: The demand and supply sides of government redistribution. Oxford Review of Economic Policy 22(3): 365–389.

11.

Dimick

Rueda

Stegmueller

(2017) The altruistic rich? Inequality and other-regarding preferences for redistribution. Quarterly Journal of Political Science 11(4): 385–439.

12.

Esping-Andersen

(1990) The Three Worlds of Welfare Capitalism. Princeton, NJ: Princeton University Press.

13.

Estévez-Abe

Kim

Y-S

(2014) Presidents, Prime Ministers and politics of care: Why Korea expanded childcare much more than Japan. Social Policy & Administration 48(6): 666–685.

14.

Garrett

(1995) Capital mobility, trade, and the domestic politics of economic policy. International Organization 49(4): 657–687.

15.

Gingrich

Ansell

(2012) Preferences in context: Micro preferences, macro contexts, and the demand for social policy. Comparative Political Studies 45(12): 1624–1654.

16.

(2008) Globalization, veto players, and welfare spending. Comparative Political Studies 41(6): 783–813.

17.

Haggard

Kaufman

(2008) Development, Democracy, and Welfare States: Latin America, East Asia, and Eastern Europe. Princeton, NJ: Princeton University Press.

18.

Hall

Soskice

(2001) Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press.

19.

Hays

Ehrlich

Peinhardt

(2005) Government spending and public support for trade in the OECD: An empirical test of the embedded liberalism thesis. International Organization 59(2): 473–494.

20.

Hwang

Lee

(2014) Globalization, factor mobility, partisanship, and compensation policies. International Studies Quarterly 58(1): 92–105.

21.

Jæger

(2009) United but divided: Welfare regimes and the level and variance in public support for redistribution. European Sociological Review 25(6): 723–737.

22.

Kang

Kim

(2010) Estimating the economic impact of the 2008 Ham-Pyeong world butterfly and insect expo Korea-using regional input-output model. Korean Journal of Local Government and Administration Studies 24(1): 61–78.

23.

Kerr

(2014) Income inequality and social preferences for redistribution and compensation differentials. Journal of Monetary Economics 66: 62–78.

24.

Kim

(2007) Openness, external risk, and volatility: Implications for the compensation hypothesis. International Organization 61(1): 181–216.

25.

Korpi

Palme

(1998) The paradox of redistribution and strategies of equality: Welfare state institutions, inequality, and poverty in the western countries. American Sociological Review 63: 661–687.

26.

Koster

(2014) Economic openness and welfare state attitudes: A multilevel study across 67 countries. International Journal of Social Welfare 23(2): 128–138.

27.

Kuhnle

(2004) Productive welfare in Korea: Moving towards a European welfare state type. In: Mishra

Kuhnle

Gilbert

, et al. (eds) Modernizing the Korean Welfare State: Toward the Productive Welfare Model. New Brunswick, NJ: Transaction Publishers, pp.47–64.

28.

Larsen

(2007) How welfare regimes generate and erode social capital: The impact of underclass phenomena. Comparative Politics 40(1): 83–101.

29.

Lee

J-S

(2011) Economic impact of Andong mask-dance festival-using regional input-output model. The Journal of the Korea Contents Association 11(1): 371–378.

30.

Lim

Burgoon

(2020) Globalization and support for unemployment spending in Asia: Do Asian citizens want to embed liberalism? Socio-Economic Review 18(2): 519–553.

31.

Lim

Tanaka

(2019) Work disincentive perceptions and welfare state attitudes: A survey experiment in South Korea. Governance 32(3): 457–473.

32.

Lim

Tanaka

(2020) The paradoxical effect of welfare knowledge: Unveiling income cleavage over attitudes to welfare in South Korea. International Political Science Review. Epub ahead of print 4 March. DOI: 10.1177/0192512120906009.

33.

Lü

Scheve

(2016) Self-centered inequity aversion and the mass politics of taxation. Comparative Political Studies 49(14): 1965–1997.

34.

Lü

Scheve

Slaughter

(2012) Inequity aversion and the international distribution of trade protection. American Journal of Political Science 56(3): 638–654.

35.

Margalit

(2011) Costly jobs: Trade-related layoffs, government compensation, and voting in us elections. American Political Science Review 105(1): 166–188.

36.

Moene

Wallerstein

(2001) Inequality, social insurance, and redistribution. American Political Science Review 95(4): 859–874.

37.

Newman

(2014) My poor friend: Financial distress in one’s social network, the perceived power of the rich, and support for redistribution. Journal of Politics 76(1): 126–138.

38.

Nooruddin

Simmons

(2009) Openness, uncertainty, and social spending: Implications for the globalization—Welfare state debate. International Studies Quarterly 53(3): 841–866.

39.

Norton

Ariely

(2011) Building a better America—One wealth quintile at a time. Perspectives on Psychological Science 6(1): 9–12.

40.

Park

B-G

(2008) Uneven development, inter-scalar tensions, and the politics of decentralization in South Korea. International Journal of Urban and Regional Research 32(1): 40–59.

41.

Potrafke

(2019) The globalisation—Welfare State Nexus: Evidence from Asia. The World Economy 42(3): 959–974.

42.

Pritchett

(2005) A lecture on the political economy of targeted safety nets. Social Protection Discussion Paper Series 501, January. Washington, DC: World Bank.

43.

Rehm

Hacker

Schlesinger

(2012) Insecure alliances: Risk, inequality, and support for the welfare state. American Political Science Review 106(2): 386–406.

44.

Rickard

(2012) Electoral systems, voters’ interests and geographic dispersion. British Journal of Political Science 42(4): 855–877.

45.

Rickard

(2015) Compensating the losers: An examination of congressional votes on trade adjustment assistance. International Interactions 41(1): 46–60.

46.

Rodrik

(1997) Has globalisation gone too far? California Management Review 39(3): 29–53.

47.

Rodrik

(1998) Why do more open economies have bigger governments? Journal of Political Economy 106(5): 997–1032.

48.

Rudra

(2002) Globalization and the decline of the welfare state in less-developed countries. International Organization 56(2): 411–445.

49.

Rudra

(2007) Welfare states in developing countries: Unique or universal? The Journal of Politics 69(2): 378–396.

50.

Rueda

Stegmueller

(2016) The externalities of inequality: Fear of crime and preferences for redistribution in Western Europe. American Journal of Political Science 60(2): 472–489.

51.

Ruggie

(1982) International regimes, transactions, and change: Embedded liberalism in the postwar economic order. International Organization 36(2): 379–415.

52.

Schaffer

Spilker

(2016) Adding another level individual responses to globalization and government welfare policies. Political Science Research and Methods 4(2): 399–426.

53.

Scheve

Slaughter

(2004) Economic insecurity and the globalization of production. American Journal of Political Science 48(4): 662–674.

54.

Scheve

Slaughter

(2001) Globalization and the Perceptions of American Worker. Washington, DC: Institute for International Economics.

55.

Schmidt-Catran

(2016) Economic inequality and public demand for redistribution: Combining cross-sectional and longitudinal evidence. Socio-Economic Review 14(1): 119–140.

56.

Schwarze

Härpfer

(2007) Are people inequality averse, and do they prefer redistribution by the state? Evidence from German longitudinal data on life satisfaction. The Journal of Socio-Economics 36(2): 233–249.

57.

Shayo

(2009) A model of social identity with an application to political economy: Nation, class, and redistribution. American Political Science Review 103(2): 147–174.

58.

Stolper

Samuelson

(1941) Protection and real wages. The Review of Economic Studies 9(1): 58–73.

59.

Yang

(2013) Parochial welfare politics and the small welfare state in South Korea. Comparative Politics 45(4): 457–475.

60.

Walter

(2010) Globalization and the welfare state: Testing the microfoundations of the compensation hypothesis. International Studies Quarterly 54(2): 403–426.

61.

Walter

(2017) Globalization and the demand-side of politics: How globalization shapes labor market risk perceptions and policy preferences. Political Science Research and Methods 5(1): 55–80.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.17 MB