Job Insecurity and Globalisation: What Factors Shape ... · Job insecurity in public discussion refers to a number of inter-related situations: actual unemployment, the expectation

Job Insecur i ty and Global isat ion: What Factors Shape Publ ic Opin ion in As ia

and Europe?

Shaun Wilson, Ian Marsh and Trevor Breusch Political Science and ACSPRI Centre for Social Research

Research School of Social Sciences

The Australian National University, Canberra

Contacts:[email protected], [email protected]

Paper for Session 16-7 Globalization and Domestic Compensation (co-sponsored by 12-24) at the Annual Meeting of the American

Political Science Association, Boston, Mass., Saturday 31 August, 1.30-3.15PM.

Job Insecurity and Globalisation 2

Abstract

Although evidence varies, the process of globalisation has produced concern among the

public in terms of its possible effects on job security. Our objective is to discover some of the

determinants of the variation in public opinion on this issue. We analyse public opinion about

the effects of globalisation on job security [using a multi-response item] based on the Asia-

Europe Study (ASES) study of 18,000 respondents in nine Asian and nine European

countries, which studies regions and globalisation. We use multinomial logistic regression

techniques to determine the factors that are significant in explaining variation in responses on

job security including: demographics, economic situation, value orientations, national

political and policy attitudes, and selected economic and migration data. Regression models

are specified for the respective regions. Important findings from the results for Europe are

that measures of economic insecurity, worries about immigration, dissatisfaction with

national government performance and political values are positively associated with ‘bad

effect’ responses. In Asia, there is strong support for the good effects of globalisation on job

security, especially among younger respondents. Importantly, this support appears to be more

widespread and is not predicted by 'winner status' (education and living standards) as is the

case for Europe. Our conclusions are: negative attitudes to the job effects of globalisation in

Europe are in part values-driven and that national government performance on

unemployment - which has been generally dismal - matters in determining opinion about this

issue. Combined with evidence from the survey, which suggests that the public prefers

national solutions to unemployment, and global solutions to other problems, we suggest that

national institutions have a significant role in mediating both real effects and opinion about

this issue.

Job Security and Globalisation 3

Job Security and Globalisation:

What Factors Shape Public Opinion in Asia and Europe?

Job insecurity in public discussion refers to a number of inter-related situations: actual

unemployment, the expectation of losing one's job in a defined time period, and as Manski

and Straub add, the expectation of not finding an equivalent job in the event of job loss

(Manski and Straub, 1995, p. 3). While economic studies are divided, the process of

globalisation has produced public concern, and especially in western countries, about its

possible negative effects on job security. Public opinion is an important dimension of the

process of globalisation: experiences and perceptions about major global changes - especially

in areas like employment and employment security - have a feedback impact on both politics

and policy.

This paper attempts to discover some of the determinants of the variation in public opinion

on this issue using a multi-response item based on the Asia-Europe Study (ASES) of 18,000

respondents in nine Asian and nine European countries.1 The first part of the paper examines

some possible explanations for opinion variation on this issue. These explanations fall into

three areas. These are: the emergence of ‘winners’ and ‘losers’ as a result of global economic

integration that drives opinion cleavages; variations in national and regional approaches to

greater economic integration; and perceptions about the broader context of globalisation

(trade, immigration and the role of the United States). On the latter, we are interested in

understanding how issues like free trade, immigration, and the role of the United States in the

post-national context might be fuelling any value-driven ‘backlash’ against globalisation.

The second part of the paper uses multinomial logistic regression techniques to test the

explanatory power of some of the factors discussed in the first part of the paper in an attempt

to explain variation in opinion in job security using the multi-response item. These include:

demographics, economic situation, value orientations, national political and policy attitudes,

and selected economic and migration data. Regression models are specified for the respective

regions. The third part of the paper discusses the results and draws out relevant policy and

political implications.

1 This paper draws on data on political cultures and globalisation collected through the Asia-Europe (ASES) survey conducted in 2000, a project led by Professor Takashi Inoguchi (University of Tokyo) and sponsored by the Ministry of Education, Cultures, Sports, Science and Technology, Tokyo (Project Number 11102000). The countries surveyed in Europe were: Sweden, Ireland, Greece, Spain, United Kingdom, Portugal, France, and Germany. The countries in Asia were: Thailand, Singapore, Japan, South Korea, Malaysia, Indonesia, the Philippines, Taiwan, and China. Samples of 1,000 were collected from each country. For details of the methodology used in the collection of the samples, email: [email protected].

mailto:[email protected]


1. Globalisation, Job Insecurity and Public Opinion

Background and definitional issues

The process of globalisation has attracted significant definitional and political controversy.

While few scholars doubt the novelty of worldwide economic, social and political integration,

some have argued that regional and national factors are ignored in over-general explanations.2

Nonetheless, the deeper integration of the world economy has by now readily identifiable

features. Change has been particularly marked over the past decade. For example, according

to the World Bank, world trade as a share of world Gross Domestic Product increased from

22.5 to 27.4 percent between 1989 and 1999. Over this same period, world trade in goods as a

share of world GDP increased from 85 to 112 percent (including re-exports). Similarly, gross

private capital flows as a share of GDP rose from 8.5 to 18.3 percent. And finally, foreign

direct investment as a share of world GDP increased from 2 to 4.6 percent. At the level of

individual states, a number in East and Southeast Asia have become significant players in

world trade, participants in global supply chains and significant recipients of FDI (EIU,

2002). Detailed estimates of the extension of trade in goods and services markets forecast

continuing and substantial increases (Fraser and Oppenheim, 1997). On this view, technology

and/or trade liberalisation will make hitherto geographically segmented markets

progressively more accessible.

These economic transformations have significance for individuals in particular societies in

relation both to their immediate material circumstances and to their perceptions and attitudes.

The economic processes identified as part of globalisation have coincided with evidence of

increasing material inequalities. Within rich countries, the story is increasingly one of

widening divisions (Therborn 2002, p. 16; Hout 2002). While nations, regions and

individuals all benefit from the opportunities opened up by global economic integration, both

political actors and scholars attribute high unemployment and/or the high incidence of

working poor to the painful downside of global adjustment. Between the first and third

2 At the outset, reservations – and qualifications - about the idea of economic globalisation should be noted. In one perspective, this argument has been linked to the national identity of multinationals and the (differentiated) dispersion of their activities (Hirst and Thompson, 1997; Hirst, 2002). Considering Europe, Fligstein argues the aggregate economic data that is presented, as evidence of ‘economic globalisation’ is better interpreted as economic ‘Europeanisation’. This is an important factual dispute whose outcome is potentially significant for public attitudes and thus relevant to the findings of this paper. But for the moment these arguments remain between scholars. In a political context, public perceptions are the relevant causal variable. The elaborated governance framework represented by the European Union does however point to another issue. In assessing the political significance of perceptions of job insecurity, we will also want to discover what standing, if any, regional governance has attained in the minds of citizens.


worlds, another dimension of inequality is revealed. While income gains both absolutely and

also in per capita terms in east and Southeast Asia - especially China - have been substantial

(Therborn 2002, p. 13), other regions like Latin America and Africa have experienced little

benefit or significant decline (Scott, 2001; Kohli et al., 2002).

Why does globalisation figure in public insecurity about employment? A first response to

this is that global processes clearly do have the potential to shift production and work and

make employment more insecure. The globalisation of production - in theory - means that

more employment becomes dependent on the contingencies of a more mobile and easily

restructured capital base, generating a new source of insecurity in employment beyond

national business and policy cycles. A second response is that global processes are perceived

to generate of insecurity because of their distance from their consequences. The association

of globalisation and insecurity is captured in the core of Giddens’ definition of globalisation:

‘(Globalisation is) the intensification of world wide social relations…in such a way that local

happenings are determined by events occurring many miles away’ (Giddens, 1990, p. 64).

Decision-making bodies at the regional and global level (European Union, ASEAN, trade

organisations and multinationals) compound this insecurity as does the widespread view that

national policy is impotent in a global economy or is now hemmed in by ‘the golden

straitjacket’ to use Friedman's popular expression (1999, p. 87).

In this section, we look at possible driving factors that might explain the variation in

attitudes about the impact of globalisation on job insecurity. These are: patterns of ‘winners’

and ‘losers’ from global economic integration; the performance of national governments and

supra-national regions in achieving global economic integration; and broader perceptions and

values about globalisation.

Winners and Losers: Is Global Trade a Factor in Job Insecurity?

Rising job insecurity in the OECD during the 1990s (and in Asia especially after 1997)

sparked debate about how much of this insecurity could be explained by globalisation and

how much could be explained by other factors. Certainly, concern about job losses from

advanced economies to third-world economies helped drive the debate about whether some in

the West were ‘losers’ from greater integration.

Defenders of globalisation have often pointed to the theoretical gains from free trade to

reject both claims about these negative employment and security effects and arguments for

greater protectionism. Their defence is based on the net efficiency gains from free trade:

distributional issues, employment and job security impacts either are assumed to follow from

an efficiency maximisation approach or are seen as independent policy goals. Kapstein cites


trade theorist and now official at the International Monetary Fund, Anne Krueger, on this

point: ‘Any effort to analyse deeper integration … must weigh proposals according to their

impact on the economic efficiency of the world' (Krueger, 1995 cited Kapstein, 2000; also

Bhagwati, 2002; Irwin, 2002).

But this approach has been questioned both in terms of the efficiency-first approach of

trade theory and in the real-world patterns of global integration.3 The reality of deeper

economic integration differs substantially from the image of a free-trade model of world

economic organisation; it consists of protectionist regimes and instruments, concentrated

economic interests and powerful and weak states. Therborn makes it clear how real-world

complexities muddy the global trade arguments:

In spite of their strong free trade instincts, mainstream economists are increasingly recognizing the ambiguous effects of trade. … What is coming out of this fairly intense economic debate might be termed a sociological insight, that trade liberalization as well as protectionism have both losers and winners and that who is who does not follow from stylised assumptions of factor endowments, but depends on the country specific institutions of the free trade regime (2002, pp. 20-21).

Detractors from the free-trade conventional wisdom belong to two groups. The first points to

direct evidence of employment losses from global integration and the second points to other

factors as more important in explaining unemployment and insecurity.

Direct evidence of employment losses in rich countries associated with global trade has

been produced in a number of studies. Kapstein (2000) has usefully summarised some of the

evidence so far. He cites two particular studies. First, Adrian Wood concluded that ‘more than

half the decline in demand for unskilled workers in the north is due to increasing

interdependence’ (cited Kapstein, p. 370). William Cline's study, which focused on the

United States, found that the combined effect of changes in trade and immigration ‘ … may

be responsible for 20-25 percent of the increase in wage inequality … [and that] this effect

will grow in the future as trade with developing countries expands’ (cited Kapstein, p. 371).

More recent studies have added to this evidence. Rowthorne and Ramaswamy (1998)

estimated that 80 percent of the fall in industrial employment in the advanced economies

3 For example, a spirited exchange between Kapstein and Krugman occurred in the pages of the journal Foreign Affairs in 1997 (Vol. 57, Nos. 3 and 4). One aspect concerned the actual contribution of economic globalisation to job insecurity. Paul Krugman, in common with other contributors, asserted the evidence did not support any significant linkage between these variables (1997, pp. 164-179). In Krugman’s view, technological change, not expanding international trade, was responsible for chronic European unemployment and growing American income inequalities.


between 1964 and 1994 could be explained by internal factors and 20 percent by competition

from low wage economies. At another level, Kate Bronfenbrenner's study suggests that even

the threat of job losses may have a far-reaching impact. She suggests that since the North

American Free Trade Agreement (NAFTA) was signed, employers threaten to close in half of

the union organising campaigns and ‘where union organizing drives are successful,

employers do in fact close their plant, in whole or in part, 15 percent of the time - triple the

pre-NAFTA rate.’(1997, pp. 1-5).

Some scholars have used the kind of evidence presented above to argue that the current

model of globalisation at its core does not promote labour's interests. Robert Wade, for

instance, argues the ultimate objective of globalisation is to shift the terms of the relationship

between states and their citizens away from any social democratic settlement. The idea that

society should shelter the individual from certain risks is being displaced in favour of

individual responsibility for risk – this is, in effect, ‘a realignment of interests away from the

protection of labour income towards the protection of capital income’ (Wade, 2001, p.27).

Heightened job insecurity is a necessary, if indirect, consequence of these broader purposes.

According to Wade, the motor of US policy is ‘what could be called the Wall Street-Treasury

complex, an analogy with the military industrial complex. It links Wall Street, the US

Treasury, the US Congress, the IMF, the World Bank, and the US economics profession …

with the City of London and the UK Treasury as an offshore adjunct. This complex is the seat

of the great campaign going on through the late 1980s and 1990s to open up the emerging

markets to free capital mobility, with the US Treasury at the helm’ (Wade, 2001, p.26)

While the losses among some sectors from global integration are widely acknowledged,

other scholars point to different factors in explaining employment outcomes that affect

worker insecurity. Certainly, rising worker insecurity in the 1990s - notably acknowledged by

Federal Reserve Chairman Alan Greenspan as a factor in the growth scenario during the

1990s - does not correlate neatly with trade flows or trade exposure. In a review article,

Taylor (2002) has defended global trade and asserted, like Paul Krugman, that technology

and not globalisation is responsible for job losses and insecurity.4 Equally, OECD evidence

suggests the level of job insecurity is not necessarily related to the level of unemployment:

4 Taylor argues, ‘The circumstantial evidence that international trade provides economic benefits is overwhelming … There is no evidence that globalisation increases overall unemployment rates … However international competition can cause reallocations of jobs from one industry to another…This problem is best described as an increase in what economists call “displaced workers”… (Public policy) should help workers make the transition to new employment.’


high levels of job insecurity in the United States and the United Kingdom persisted amidst

low and falling official unemployment (OECD, 1997).

A further perspective on the causes of job insecurity and widening income gaps

acknowledges the role of economic integration but underscores the contribution of state

strategies. Fritz Scharpf, while acknowledging the effects of globalisation on low-skilled

employment in the first world, goes on to argue that ‘overall employment levels are not

primarily determined by factors affecting the competitiveness in the internationally exposed

sectors of the economy’ (1997, p. 2). Scharpf sees ‘high employment’ societies like the

contrasting models of Sweden and the United States as promoting employment in the post-

national era through non-exposed services; in the former case, employment has grown in the

state services sector and in the latter, employment has grown in low paid private services.

Scharpf clearly thinks that employment levels - and by extension - employment security - are

problems that persist for national governments even under conditions of global integration.

He sees, however, difficulties in replicating either the Swedish or American solutions,

namely, tax resistance in the former and unacceptable inequality in the latter.

T A B L E O N E : G L O B A L I S A T I O N A N D J O B S E C U R I T Y : A T T I T U D E S B Y S E L F - R E P O R T E D L I V I N G S T A N D A R D S A N D

R E G I O N

Asia (in percent) Living Standards Good Bad Effect No effect Don't know

High 38 17 29 11 6 Relatively high 32 13 30 16 8

Average 32 16 27 17 8 Relatively low 26 19 30 15 10

Low 23 21 24 16 17

Europe (in percent)

Living Standards Good Bad Effect No effect Don't know High 34 20 24 17 5

Relatively high 28 16 25 24 7 Average 24 22 26 21 8

Relatively low 18 29 19 22 12 Low 15 25 13 33 14

ASES questions: Please tell me whether [globalisation] has any effect on your own life in [terms of job security] and whether it is a good thing or a bad thing; ‘how would you describe your household’s living standards’.

Evidence particularly points to some employment losses among first world economies

from global integration especially in low skill, low wage sectors exposed to economic

competition. No doubt major relocations of factories and job losses in old industries influence


first world public opinion about the effects of globalisation, and probably more so than the

gains from integration that accrue to the ‘winners’ (for US evidence, see Scheve and

Slaughter, 2002).

Evidence compiled from the ASES data gives some insight into the perceptions of

‘winners’ and ‘losers’ about the effects of globalisation on job security. When responses to

the question about the effect of globalisation on respondent job security were categorised

according to self-reported living standards, some clear patterns in Asia and Europe emerge

(see Table One). While both Asians and Europeans with high self-reported living standards

were relatively positive about the effects of globalisation on job security, regional divergence

is evident among respondents with lower living standards. Low-income respondents in

Europe on balance held unfavourable views (the ‘bad effect’ responses outnumbered the

‘good effect’ responses by around 10 percent) while low-income respondents in Asia held

more favourable views (the good effects still outnumbered the bad effects but by a smaller

margin than higher income Asian respondents). This may be suggestive of a ‘globalisation

backlash’ among low-income households in Europe where realities and perceptions of job

losses from economic integration coincide.

National and Regional Performance: Mediating the effects of global integration?

Scharpf's observations about the importance of national structures in determining

employment levels in the post-national era serves as a reminder of the importance of national

policies and institutions in mediating the process of global integration. But do the same

national policy institutions make a difference to the reception of globalisation by their

respective national publics? To answer this, we must consider the way national policies act to

co-ordinate national and global economic outcomes and compensate those who are affected

by global exposure, and (potentially) redistribute the gains. A significant literature, mainly

interested in the European context, has concentrated in the types of redistributive action

undertaken by national welfare states and the policy, politics and actors that support and

distinguish it (see Esping-Andersen, 1990). Further work has shown that welfare state

policies become the institutional basis for public opinion about legitimate public policy (see

Svallfors, 1997; Pierson, 1994). For example, in strongly redistributive welfare states, support

for redistribution is generally higher than in market or liberal states. Less work has been done

on the institutional basis - particularly its impact on public opinion - of various models of

national-global interaction through trade. Our interest is to investigate whether the same

holds true for attitudes towards the employment effects of globalisation: do successful ‘export

models’ encourage pro-globalisation attitudes when it comes to employment?


Some European states included in the ASES survey have very successful export-led

economies with high export/GDP ratios. It is significant that respondents in Sweden and

Ireland (see Table Two) hold the most positive views about globalisation on this score of the

European countries surveyed; for different reasons, both have benefited from global

openness, exports and inward foreign investment. Other countries have similar export/GDP

ratios and similar range of opinion on globalisation, so it is worth commenting on the features

of the two outliers.

T A B L E T W O : E X P O R T S A N D A T T I T U D E S : E F F E C T O F G L O B A L I S A T I O N O N J O B S E C U R I T Y B Y C O U N T R Y

Europe (figures in percent)

Good Bad Effect No effect Good -Bad Export/GDP

(percent) IT 24 39 28 10 -13 26

SP 17 22 36 25 -5 27 UK 19 24 28 29 -5 26 PO 25 28 20 26 -3 30 GR 35 34 20 11 1 20 GE 25 24 31 19 1 29 FR 27 24 26 23 3 26 SW 22 11 31 36 11 47 IRE 37 13 18 32 24 88

Asia (figures in percent)

Good Bad Effect No effect Good -Bad Export/GDP

(percent) SK 25 31 33 11 -6 42 JA 8 13 43 36 -5 10

CH 28 24 26 23 4 22 TA 35 23 26 16 12 na TH 41 24 22 14 17 58 PH 36 18 37 10 18 51 SI 36 16 35 13 20 166

MA 46 9 24 21 37 122 IN 45 7 29 20 38 35

ASES question: Please tell me whether [globalisation] has any effect on your own life in [terms of job security] and whether it is a good thing or a bad thing … Export/GDP data for countries obtained from World Bank (2002).

The Swedish model – long thought to be incapable of sustaining itself in post-national

contexts – has survived largely intact (Stephens, 2000) with a politics of domestic

compensation with strong labour interests accepting the free-trade export model on the basis

of state redistribution and a solidaristic wages policy (Meidner, 1993). Sweden’s recent

export performance – advantaged by a currency devaluation and significant new investment

in information technology industries – has stimulated export growth and employment. It


would appear from evidence presented in Table Two that the Swedish public accepts the

benefits of global trade – consistent with the institutional basis of the Swedish model. Ireland,

on the other hand, has benefited from very strong economic growth during the 1990s and has

undergone substantial economic development, attracting inward foreign investment in new

industries and strong export growth. Its public is the most positive about the effects of

globalisation on job security.

Asian respondents are generally much more positive about the effects of globalisation on

job security than Europeans taken as a whole. This again may reflect the relative benefits of

recent export growth. It may also reflect the relative stage of development where the losses

from global integration at the hands of lower-cost, higher productivity competitors have not

yet emerged in any general way. Scholars have drawn attention to the institutions that

underscored the ‘developmental’ approach to economic governance (see Woo-Cummings,

1999). In particular, Campos and Root (1996) have discussed the strategies of ‘shared

growth’ that they find to have been an essential element of the approach of the Asian states.

All Asian states have adopted export-led development strategies. Japan, Korea, Taiwan

and Singapore have been regarded as classic ‘developmental’ states, which involve state

leadership in the development of the economy and international engagement. The legitimacy

of the model was in part derived from the model of ‘shared growth’ that underpinned

development and at least partly from an elite consensus about the priority for export-led

growth.5

Asian states remain underdeveloped in welfare state institutions (for Southeast Asian

evidence see Ramesh, 2000). Where social institutions have developed, like the prominent

case of Japan, welfare has been implicitly tied to work status, which in turn depended on

growth to ensure full employment levels. The lack of adequate welfare institutions is

reflected in the extent to which these developing economies and societies have encountered

institutional and distributional problems in the aftermath of the Asian economic crisis of the

late 1990s, which heavily impacted on domestic growth and exports. In South Korea,

particularly, a ‘European’ level of negative opinion about the security effects of globalisation

is evident in Table Three, reflecting perhaps serious discontent and institutional inadequacies

that have only grown with the hardships of the post-1997 economic crisis.

5 As Kim explains in the case of Japan: ‘the idea of catching up and surpassing the West (oitsuke, oikose) was generally accepted throughout the population, including by the elite of the ruling party…It was the pervasive anxiety arising from fuan (insecurity) that helped mobilise the people…The state adopted the principle of shared growth’ (Kim et al. 1995, p. 515).


T A B L E T H R E E : T H E E C O N O M Y A N D U N E M P L O Y M E N T : G L O B A L O R N A T I O N A L P R O B L E M S ?

Problems in the Economy Unemployment

Causes of …. (in percent) National Global Both National Global Both Europe Greece 25 21 49 42 15 38 Spain 39 29 27 63 16 16 Portugal 31 22 39 50 13 31 Germany 27 29 39 58 14 27 Italy 33 19 43 56 14 27 France 9 41 45 34 25 38 United Kingdom 31 26 35 60 14 21 Sweden 24 25 47 52 16 30 Ireland 35 30 27 63 23 18 Asia Philippines 36 38 25 54 28 17 South Korea 50 12 34 66 8 22 Japan 26 8 52 55 4 30 China 42 11 47 74 5 20 Taiwan 40 13 44 58 8 32 Thailand 30 18 48 45 13 38 Indonesia 49 10 37 79 3 16 Malaysia 32 19 44 58 9 28 Singapore 9 36 50 27 24 44 ASES question: Do you feel that the [problems in the economy, unemployment] in [country] are due mainly to causes within [country] or are due to causes in the international situation, or both? Given the background of welfare and developmental state policy and its potential for

mediating and co-ordinating the effects of global change, has the public in either region

shifted its expectations away from national governments and towards more globally co-

ordinated policies? ASES evidence establishes important trends. When respondents were

asked whether the causes of problems in both the economy and unemployment were due to

national or global factors, or both equally, a consistent divergence in opinion emerged (see

Table Three). In Europe, the modal response for the question on the causes of ‘problems in

the economy’ for all countries except Spain was ‘both equally’. In contrast, the modal

response to the question on the causes of unemployment in all European countries except

France was ‘national’ causes. In Asia, the modal response in six countries was ‘both equally’

for the question on the economy and ‘national’ for all countries for the question on

unemployment. Taken together, these findings suggest that the public in Asia and Europe

attach global causes to economic problems as much as they do national causes but in both

regions, the problem of unemployment is considered to have primarily national causes.


TABLE F OUR: J O BS AS A NAT IONA L PROBL EM IN A GLOB AL

WORLD?

(in percent) Human Rights Environment Unemployment Problems dealt with by Country All C'tries Country All C'tries Country All C'tries

Europe Greece 32 63 33 62 54 44

Spain 17 74 22 71 59 34 Portugal 14 78 26 68 45 50

Germany 20 73 15 81 60 36 Italy 16 77 24 72 44 52

France 18 79 21 76 45 52 United Kingdom 31 60 26 68 76 19

Sweden 17 70 11 85 65 29 Ireland 36 57 46 48 70 25

Asia

Philippines 32 65 30 67 37 59 South Korea 28 62 25 69 63 29

Japan 17 59 11 76 51 32 China - - 43 52 76 16

Taiwan 63 27 75 20 81 13 Thailand 50 38 60 35 69 27

Indonesia 65 25 71 25 77 21 Malaysia 52 33 55 37 70 24

Singapore 40 42 36 58 61 33 NB. Human rights question was not asked in China. ASES question: Would you please tell me whether [Problems of human rights, Environmental problems, The problem of unemployment] should be dealt with by each country deciding for itself what should be done or by all countries together deciding what should be done… This finding is placed in larger context when responses to the question about whether

various problems should be dealt with by each country deciding for itself or by all countries

together (see Table Four). Respondents were asked about a range of problems including

human rights, asylum seekers, environmental problems and unemployment. A sharp contrast

in responses emerges on the problem of unemployment when compared to other problems

like the environment and human rights. In Europe, respondents in all countries were on

balance in favour of all-country solutions to human rights and environmental problems but, in

six of the nine countries, national responses were preferred for unemployment. In Asia,

national responses for human rights and environmental problem were preferred in more cases

than in Europe but, again, the strongest preference for national policies was elicited for the

question on unemployment. Solutions to the problem of unemployment require – in the

public's mind at least - significant national involvement and intervention, and this remains so


even where other problems and policy areas are now considered to be better solved by a co-

ordinated global response. While the public in both regions generally accept the economic

consequences of global integration, they still think employment problems require national

responses. One interpretation of this is that the public expects - especially with regard to

economic security - national governments to play a mediating and compensating role in the

post-national era.

Public opinion and the broader context of globalisation: trade, immigration and the United

States

While individual material effects and national and regional policies and outcomes may

help to explain public opinion, attitudes on globalisation are not reached only on the basis of

national or individual material circumstances. This is in part because global processes reach

beyond trade flows. Clearly, attitudes towards global institutions, migration, and issues and

problems that cross national borders like human rights and the environment inform overall

and specific opinions. Here, we concentrate on three particular political issues that are

attached to increased global integration: free trade and protectionism, migration and the

global influence on the United States. These issues have been identified as part of the

growing ‘politics of globalisation’ that increasingly shapes debates and conflicts, and possible

‘backlashes’ against global change. Commentators have raised the prospect of a protectionist

backlash in public opinion - especially in rich countries - as a result of insecurities generated

by global integration; far-right parties in a number of countries have exploited anti-

immigration feelings in Europe and the global role of the United States has come into greater

focus perhaps especially since the terrorist attacks on September 11, 2001.

The persistence of protectionist sentiment in spite of pre-dominance of free trade views

around the world may be evidence of public resistance towards economic globalisation. In his

study, Richard Rose found evidence of wide popular support for protection in a study that

covered twenty countries including ten of the eighteen included in the present survey. Forty

six per cent of respondents supported protection of domestic industries and 45 per cent

supported free trade with 9 percent undecided (Rose, 2002, p.1). Rose described his

‘protectionist’ respondents as ‘confused’. This is because he found majorities in this same

group who also liked the new abundance and variety of goods in shops. But this judgment

would seem to privilege the value of consumption over employment or broader political

preferences.

It is plausible that respondents understand adjustment processes can be very costly for

particular groups, including themselves, and that they want government to minimise these


impacts. This is a conclusion reached by other scholars who have generally confirmed that

protectionist sentiment is strongest among low skilled workers (Scheve and Slaughter, 2001;

Mayda and Rodrik, 2002) and those with low incomes (Mayda and Rodrik, 2002).6 In later

work focussing on the perceptions of American workers generally about the costs and

benefits of globalisation, Scheve and Slaughter (2002) found that:

The majority ... have concerns about these (international) transactions, in particular their adverse labour market impacts. On balance, more people seem to weight these costs as more important than the benefits, such that…they respond with preferences for policies aimed at less not more liberalisation of trade, immigration, and FDI…Less skilled individuals, measured by educational attainment or wages earned, are much more likely to oppose freer trade and immigration than their more skilled counterparts’ (cited Swank and Betz, p. 12).

Meantime, Rodrik traces the likely political consequences arising from a process of

globalisation that is sharply divided between winners and losers:

What is perhaps the greatest risk of all, namely that the cumulative consequences of (globalisation) will be the solidifying of a new set of class divisions between those who prosper in the globalised economy and those who do not, between those who share its values and those who would rather not, and between those who can diversify away its risks and those who cannot. This is not a pleasing prospect, even for individuals on the winning side…social disintegration is not a spectator sport – those on the sidelines also get splashed with mud from the field. Ultimately, the deepening of social fissures can harm all (p. 33). The opinion of the apparent losers from globalisation is certainly one of many factors

explaining the further entry and success of far right populist parties in Europe, which have all

campaigned on the issue of immigration.7 Duane Swank and Hans-Georg Betz (2002) have

analysed support for populist right wing parties in Europe. Swank and Betz assess the extent

to which the support base of right wing populist parties is associated with ‘groups that lose

from contemporary features of modernisation’ (p.1).

Swank and Betz argue ‘at comparable levels of taxation, comprehensive generous and

employment oriented welfare states do a better job than other systems of social protection in

6In a US context, Scheve and Slaughter (2001) have explored the influence of employment, skill levels and asset ownership on trade policy preferences. They found that the skill level of workers is a better predictor of individual support for restrictions on trade than employment in a trade-exposed industry. They also found that ‘home ownership in counties with a manufacturing mix concentrated in comparative-disadvantage industries is strongly correlated with support for trade barriers’ (p. 269-270). However, Mayda and Rodrik's twenty three-country study concludes that the sector of employment was associated with policy preferences, with individuals in non-tradable sectors exhibiting the most pro-trade views and protectionist sentiment associated with trade exposure (2002). 7 All nine European countries covered in this study are included in their sample that involves all elections from 1981 to 1998. Right wing populist parties were particularly successful in two states: France and Italy between these periods and in the subsequent elections in both countries not covered in the study held in 2002 and 2001 respectively. On the failure of right wing populism in Sweden, see


buffering individuals from risks to income and employment posed by internationalisation and

bolstering support for state institutions’ (2002, p. 2). In reaching these judgements, they

tested the links between voting outcomes and a variety of ‘real’ changes. They found that

trade openness and capital mobility were not directly associated with vote for right populist

parties (p. 31). However, the volume of refugees and asylum seekers was systematically and

positively related (p. 31). The direct effect of welfare state structures was also highly

significant: universalistic welfare states significantly depressed the vote of populist right

parties. Deindustrialisation, proportional representation, past electoral success of left-

libertarian parties and average effective tax burdens were also significantly associated with

greater electoral success for these parties (p. 32).

ASES evidence sheds some further light on this. Our findings suggests that - at least in

Europe where immigration concerns are, currently, highly electorally significant -

respondents holding negative views about the job security effects of globalisation also tended

to be more worried about immigration. This is also sharply apparent for country-by-country

findings, which are reported in Figure 1, which plots the number of respondents who were

‘very worried’ about immigration against the number of respondents who thought

globalisation had a ‘bad effect’ on job security for the nine European countries. Only

Portugal (which has the lowest foreign born labour force of the nine countries) and Sweden

(which has a lower level of concern about globalisation's effect on job security) differ from

the general trend. One is left to speculate about the relationship between concerns about

globalisation, job security, and immigration. Fears about immigrants may be part of a

worldview that is hostile to globalisation because it is perceived that new immigrants take

jobs. High levels of unemployment may drive concerns about the level of immigration. These

insecurities may be reinforced in neighbourhoods where large numbers of unemployed

migrants may fuel perceptions that the problems of economic insecurity and immigration are

automatically linked.

Rydgren (2002).


FIGURE ONE

Job Insecurity and Worries about Immigration - Europe

0

5

10

15

20

25

30

35

40

0 20 40 60

"Very worried" about immigration

Glo

balis

atio

n ha

s "b

ad e

ffect

s"

on jo

b se

curit

y

InsecurityLinear (Insecurity)

Perceptions of globalisation that have been driven by dominant economic and cultural

forces focus attention on the role of the United States in this process. Many commentators

have argued that the global model is driven by ‘Americanisation’ - the imposition of a US

model of economics and culture on the rest of the world through its multinational, military,

and institutional interventions (see, for example, Fligstein, 1997; Wade, 2001; Chomsky,

2001; Ali, 2002). Perceptions of globalisation as Americanisation have many roots - in left

and right wing identifications - and in national-cultural and religious ideological

frameworks. When it comes to economics, is there a general perception that the American

version of capitalism carries with it greater job insecurity? ASES evidence is suggestive of

such a link. Respondents were asked whether or not it is a fair thing to describe international

influence as a form of Americanisation. Regional responses differ: Asians are generally less

likely to think it was fair to say that international influence is a form of Americanisation than

Europeans. South Korea recorded the highest number of ‘fair’ responses (61 percent) in Asia

while Greece and France - with long standing anti-American traditions - recorded the highest

results in Europe (70 percent and 60 percent respectively).


T A B L E F I V E : G L O B A L I S A T I O N A S A M E R I C A N I S A T I O N A N D

J O B S E C U R I T Y

Asia (Job security) Good Bad Effect No effect

Fair 40 41 37 35 Not Fair 40 40 40 35

Don't Know 20 19 23 30

Europe Fair 51 59 49 47

Not Fair 35 27 30 28 Don't Know 14 15 21 25

ASES questions: Some people say that all this international influence is really a form of Americanisation. Do you think this is a fair description?; Please tell me whether [globalisation] has any effect on your own life in [terms of job security] and whether it is a good thing or a bad thing …

Table Five compares responses to the Americanisation question with responses to the

globalisation and job security question. Little difference in responses in evident in the Asian

sample. The highest incidence of ‘fair’ responses is recorded for Europeans who think

globalisation has a bad effect on job security (59 percent). This may indicate that perceptions

of the American model becoming a ‘world model’ could be a driving factor in identifying

globalisation with economic insecurity.

In this section, we have presented some evidence of a ‘globalisation backlash’ on

questions of protectionism, immigration and the role of the United States in global

integration. We have identified material, national and value-based factors that seem to

distinguish attitudes about the effects of globalisation on job security, which require further

testing and evaluation. This is undertaken with the use of multinomial logistic regression

techniques in the next section.

2. Job Security and Globalisation: Regression analysis on the ASES data

Multinomial logistic regression models were specified for the ASES sample in attempt to

test the explanatory power of a number of variables that emerged as significant in the first

part of the paper. Two regression models for Europe and Asia respectively are presented here.

The dependent variable was the question ‘Please tell me whether [globalisation] has any

effect on your own life in terms of job security and whether it is a good thing or a bad thing

…’, which is coded as JOBSEC. This allowed comparison of those respondents who thought


globalisation was having a ‘good effect’ and those who thought it was having a ‘bad effect’,

"an effect but neither good nor bad’, ‘no effect’, and those who responded ‘don't know’.

Independent variables were entered in blocks and the models were respecified after omitting

variables with large numbers of missing cases (which reduced sample size) and low overall

significance. For variable descriptions and details of the variable coding in the respective

models, refer to Appendix A.

2. Regression model A: Europe

The first block of variables deals with some background information about the

respondents: age, gender, years of schooling and a question about the respondent’s

employment situation. Years of schooling are taken to be a measure for the level of education

of the respondent and coded as SCHOOL. [At this stage, the ASES data on educational

attainment includes nation-specific levels of education for which a common code is being

developed.] The question on employment situation asked if the respondent, in his/her

employment, has contact with people or organisations in other countries and coded as

GLOBALJOB. This is a measure of the integration of the respondent's workplace into a

regional or global environment. In Europe, this measure is quite high and, naturally, could

include long-distance truck drivers as well as bankers and academics.

In the first section of the paper, the influence of economic circumstances was discussed as

a possible factor in shaping attitudes about globalisation and job security (i.e. ‘winners’ and

‘losers’). The second block of variables therefore included two measures of subjective

economic security, which provide an indication of the respondent's level of concern about his

or her welfare. The first question asked the respondent how worried he/she was about his/her

work situation and was coded as WORKINS. This question referred to concern about work at

the most general level and can only be interpreted as such: it is not, taken alone, a specific

measure of job insecurity (in the sense of losing one's job or having one's pay or hours cut).

The second question asked the respondent to indicate his/her household income according to

the following categories: ‘high’, ‘relatively high’, ‘average’, ‘relatively low’ and ‘low’, and

was coded LIVSTAN. This measure has been used instead of household incomes reported in

national currencies for two reasons. The first is the ASES data did not include a common

scale of income levels for international comparisons and second, from the point of view of

making meaningful comparisons of economic security, a subjective assessment made

according to national relativities is sufficient for our comparative purposes.

The third block of variables includes measures of basic political orientations including

self-reported position on a ten point left-right scale commonly used in political science


(coded as LEFTRT), attitudes towards competition (COMPGOOD), level of national pride

(NATPRIDE), and attitudes on whether or not globalisation is a form of Americanisation

(AMERICA). These political or ideological orientations - taken individually or as a block -

help separate out ‘material’ and ‘value’ influences shaping opinion on job security and

globalisation, and which we suggested in the previous section may be implicated in the

political backlash against globalisation.

The fourth block of variables measures the respondent's attitudes towards national

problems and government policies. We are primarily interested to know if the national policy

situation serves as a basic frame of reference in making assessments about global impacts on

employment, and which was discussed in the previous section. Three policy problem areas

were included: concerns about the economy (WORRYEC), unemployment (WORRYUE)

and immigration (WORRYIM). Measures of respondent confidence in the national

government's handling of these three problems were also included (coded as GOVEC,

GOVUE and GOVIM respectively). Three questions were included to establish levels of

confidence in political parties (CONFPOLP), political leadership (CONFPOLL) and big

business (CONFBBUS). Another question - which asked respondents to indicate the extent to

which the national government had a responsibility to the welfare and employment of its

citizens – was included and coded as GOVRES. A final question, which asked respondents

whether or not they agreed with the limitation of imports was included, and coded as

LIMIMP. This was included in order to see if there was any association between ‘bad effect’

responses (on the job security and globalisation question) and preferring import limits –

something that might be expected given the results of other studies in the area documented

previously.

The fifth block of variables measures the impact of a variety of institutions on the

everyday life of respondents. We included responses about national governments

(EFFECTGOV), the European Union (EFFECTEU), the United Nations (EFFECTUN) and

multinationals (EFFECTMN). Again, these measures may indicate which institutions explain

response variations in the dependent variable.

Finally, some basic national economic and population indicators8 were included to test the

importance of national economic performance as an explanatory factor, especially whether

the high-export/ low unemployment economies in Europe were less likely to associate

globalisation with perceived job insecurity. These were: a measure of long-term GDP growth

8 For details on statistical sources, refer to notes on each variable in Appendix A.


for 1990-2000 (LTGR), national export/GDP ratios for 1999 (EXPGDP), and unemployment

rates for 2000 (UNEMP). To further test the influence of immigration on opinion we included

two statistics: a five-year (1995-2000) net migration rate (NETMIG) and the percentage of

the labour force who were foreign born (FORPOP) in 1999.

Many of these variables might be predicted to have similar effects on the dependent

variable, so not all of them will be statistically significant contributors to an explanation of

the responses. Statistical significance was measured against a benchmark level of 1 percent,

(deemed appropriate given the large sample), although more stringent criteria for retention in

the model were adopted in the case of variables with large numbers of missing values. Given

the multinomial nature of the response variable, a distinction can be made between the

individual importance of an explanatory variable in distinguishing one particular response

(eg. 'bad effect') from the base of 'good effect' and the overall importance of the variable to

the model. Estimation results for the final model, with insignificant variables removed, are

reported in Appendix B1.

Results

Among the demographic and background variables, age and gender were overall significant

but played little role in explaining differences in responses. The exception that that older

respondents were more likely to respond that globalisation has 'no effect' on their job security

compared with the base response of 'good effect'. This is either because they were not

employed or were more likely to work in parts of the economy unaffected by global

processes. Schooling was overall insignificant and was removed from the model. However,

the inadequacies of this measure for education (which can be improved and retested in later

modelling) or the education effects being captured by other variables (living standards, for

instance) might help explain this result.9 Having contact with people or organisations in other

countries in one's work was overall significant and marginally significant in distinguishing

good and bad responses.

Both measures of economic insecurity - self-reported living standards and concerns about

job situation - were overall significant. Respondents with lower self-reported living standards

and greater worries about their work situation were more likely to think globalisation had a

bad effect on job security than those who thought it had a good effect. WORKINS was a

strong predictor of bad responses to the question. While concern about work at a very general

level did increase the likelihood of bad effect responses, bivariate analysis of the response

9 It is also relevant that skill levels predict responses on protectionism (see Mayda and Rodrik 2002,


patterns for those who were ‘very worried’ about their work situation revealed persistent

response variation (i.e. not all worried respondents thought globalisation was the cause of

their insecurity). Respondents who reported ‘no effect’ were less likely to be worried about

their work situation and were more likely than the base to report lower living standards.

The measures of political orientation were overall significant. National pride was

significantly lower among those nominating the 'bad effect' response than the base. This

result may seem, at first, counter-intuitive. But national pride may be indicative of one's

‘stake in the system’; in other words, we may be more likely to view our nationality

positively if we derive some tangible benefit from this attachment or, perhaps, less likely to

express identity-attachments if our basic circumstances like work are insecure. The left-right

scale was only marginally overall significant and omitted because of the high number of

missing cases (1640 from 9093).

Only those who nominated the 'bad effect' response were more likely to think globalisation

was a form of Americanisation than the base. By contrast, those nominating the 'good effect'

response were more likely to hold favourable views about competition than the other

response groups. This suggests that ideological differences and background opinions are

significant factors in shaping attitudes to the effect of globalisation on job security.

Measures of national problems and government policies also played an important part in

explaining difference in responses. For the three areas of national concern - the economy,

unemployment and immigration - some significant variations were apparent. We mention two

of the important ones. Worries about immigration increased the likelihood of nominating a

'bad effect' - an important finding, which we take up later. Those reporting 'no effect' on job

security were significantly less likely to be concerned about the economy, probably reflecting

their relative economic security. When it came to the government's handling of the three

issues, only the handling of unemployment was overall significant, and the other two were

omitted from the final model. Large differences between response categories emerged on

how respondents viewed the national government's handling of unemployment. Those

nominating the 'good effect' response were much more likely than the other response groups

to be satisfied with the national government's handling of unemployment. The variable

measuring attitudes towards limiting imports, LIMIMP, was not overall significant and it was

dropped from the model. No relationship between a preference for limiting imports and ‘bad

and Scheve and Slaughter, 2001)


effect’ responses was evident. None of the confidence measures (CONFPOLL, CONFPOLP

or CONBBUS) were overall significant and they were also omitted from the final model.

Measures of the impact of various institutions on respondents’ everyday lives reveal

interesting results. A contrast emerges between the good and bad effect responses. Those

nominating the 'bad effect' response were more likely to think national government had an

impact on everyday life while those nominating the good effects were more likely to think

that the European Union had an impact. This result suggests that those who hold positive

views of globalisation also recognise the impact of larger institutions while those holding

negative views, for whatever reason, feel the effects of immediate, national ones. The effect

of multinational companies was significant for the overall sample, with those nominating the

'no effect' response less likely than the other groups to report an impact. This indicates that

the impact of multinationals is significant for the three responses groups who reported that

globalisation has an effect on job security.

Economic and population indicators were also significant across the model but not always

for the respective response groups. There was a strong correlation between long term growth

and export to GDP ratios (0.92), in part driven by the Irish data.10 Of significance was the

finding on net migration: the 'good effect' response group differs from all the other responses,

suggesting that respondents from countries with high net migration rates are more likely to

report the good effects. Although Ireland has an impact on cross-national results, which had a

high incidence of good effect responses and the highest five-year average net migration rate

(4.9 per 1000), this result holds when the Irish data was omitted from the model.

Regression Model B: Asia

The model specified for the Asian sample tested the same variables (in the same blocks) as

the European sample. The results for the final model are reported in Appendix B2. Again,

variables that were insignificant or had a high number of missing data were omitted and the

model was respecified and tested. Variations in the model specification are reported below. In

the first block, the educational attainment variable, SCHOOL, was overall significant and

kept in the model. In the second block, the measure of living standards, LIVSTAN and the

variable GLOBALJOB (work contact with organisations and people in other countries) were

not overall significant and omitted from the model. In the third block, the variables measuring

10 Sweden and Ireland were identified as outliers in the previous section; the former with a strong export sector and a consistent history of policies of domestic compensation which are arguable consistent with buffering global shocks. Ireland has gone through a phase of massive GDP and export growth, which has no doubt transformed public opinion in that country. These two outliers no doubt have an impact on the results.


attitudes towards Americanisation, AMERICA and left-right orientations, LEFTRT, were

overall insignificant and omitted from the model.

In fourth block, variables measuring worries about the economy and immigration

(WORRYEC and WORRYIM) were not significant and omitted from the model. Only one of

the confidence variables [confidence in political leadership (CONFPOLL)] was included; the

measures of confidence in big business (CONFBBUS) and political parties (CONPOLP)

were omitted due to overall insignificance. Again, the variable measuring attitudes on

limiting imports, LIMIMP, was not significant and also omitted. Additionally, the variables

GOVRES and GOVUE were not overall significant and also omitted. The variable

EFFECTEU – measuring the impact of the European Union – was replaced with EFFECTAS,

which measures the impact of ASEAN on the lives of respondents. In the last block dealing

with economic and population indicators, FORPOP was not included due to the unavailability

of foreign labour force statistics for the Asian countries in the model.

Results

Those nominating the 'good effect' response were less likely to be older than each of the

other response groups, which suggests younger Asians drive the comparatively positive

responses about globalisation evidenced in the sample. A very interesting finding relates to

schooling that was not apparent in Europe. Those nominating the good effects were more

likely to have lower number of years in school than those nominating any of the other

responses, which suggests that positive views of globalisation are associated with lower

levels of school of education - perhaps a contradictory finding. However, this may be

indicative of the broad acceptance of the global economy as a source of economic wealth

creation through exports - consistent with the Asian developmental model.

Those nominating the 'good effect' response were much less likely to be worried about

their work situation than those nominating either bad or ‘effect but neither good nor bad’ but

more likely to be worried than respondents who responded ‘no effect’. On value-based

responses, Asian respondents nominating the good effects were more likely to be proud of

their national identity and hold positive views of competition than the three other response

categories. Again, having ‘a stake in the system’, as we postulated for the European results,

may explain this finding. Attitudes towards Americanisation - unlike in Europe - did not

appear to explain the 'bad effect' response to the question.

Some interesting results were reported from confidence in government policy. Those

nominating the good effects were more likely than the other four groups to be satisfied with

the government's handling of either the economy or immigration. The variable that measures


the government’s handling of the problem of immigration was overall significant but was

recoded (as GOVIM2) to deal with a large number of missing cases that compromised the

sample size.11 This indicates that cleavages in opinion about immigration may predict

responses in Asian sample as well, but in the Asian model, it is differences in opinion over

the government’s handling of the issue rather than the level of worry about immigration that

is the stronger predictor. Moreover, those nominating good effects were more likely to come

from countries with lower net migration, suggesting that immigration issues help explain

variation in opinion about globalisation and job security in Asia as well as Europe.

The effects of national governments, ASEAN and multinationals were all significant in

the model. The strongest differences between responses for these variables were recorded for

the ‘no effect’ group, which were less likely to report any impact of these institutions. Taken

with another finding, that the ‘no effect’ respondents were more likely to live in countries

with slower economic growth, these findings suggest that faster growth and greater

integration have definite an impact on public opinion in Asia.

3. Discussion of the findings and regression results

This study has explored three specific aspects of the perceived links between economic

globalisation and job insecurity; first, variations in responses across the eighteen countries;

second, the factors - regional, national, international, attitudinal, socio-demographic - that

help explain these perceptions; and third, the implications of these findings for the politics of

job insecurity and globalisation. Responses on this issue vary widely by region and by

country. There were significant differences between Asian and European responses. There

was approximately 7 percent difference between the two regions in judgements about the

positive impact of globalisation on job security (33 percent Asia compared to 26 percent in

Europe). Conversely, 6 percent more European respondents identified a negative effect

compared to Asians (24 percent compared to 18 percent). There were also relatively large

numbers of respondents in both regions (around 30 percent) who thought globalisation had an

effect but were ambivalent about the ultimate direction.

The variation in country results suggests national factors are particularly significant. In

Asia, only 8 percent of Japanese respondents judged globalisation to have good effects on job

security (regional average 33 percent). By contrast, 46 percent of Malaysian, 45 percent of

11 GOVIM2 was recoded so that the very high number of ‘Don’t Know' responses were coded with ‘Not so well’ and ‘Not well at all’. The categories ‘Very well’ and ‘Well” were recoded into one category as well, creating a binomial variable which tested whether or not respondents thought that the government handled immigration well or not.


Thai and 41 percent of Philippine respondents offered positive assessments. Similarly, only 7

percent of Indonesian (Javanese) respondents offered negative assessments (versus a regional

average of 18 percent). South Korean respondents offered the highest number of negative

assessments (31 percent versus regional average of 18 percent). Japanese respondents were

grouped in the ‘neither good nor bad’ (43 percent versus a regional average of 30 percent) or

‘no effect’ categories (36 percent versus a regional average of 18 percent). Anecdotal

evidence suggests this may reflect discomfort with the intrusiveness and anonymity of the

survey process. Equally, it may reflect the relatively low level of international engagement of

the Japanese economy despite its export success.

Amongst European respondents, Irish participants were significantly more likely to offer

positive judgements (37 percent versus a regional average of 26 percent). Other positive

responses grouped within ten percent either way of the regional average. As noted earlier,

amongst negative views, Irish (13 percent) and Swedish (11 percent) respondents occupied

the lowest pole in Europe (regional average 24 percent) with Italians (39 percent) and Greeks

(34 percent) occupying the other pole.

In sum, job security is clearly the most sensitive issue with Europeans with about one in

four respondents judging the impact of globalisation to be negative. By contrast, Asian results

suggest respondents have more favourable views about the security effects of globalisation.

Italy and Greece are the two states with the highest levels of negative respondents in Europe

and Korea has the highest level of negative respondents in Asia.

In explaining these outcomes, the findings suggest regional, national, individual and

attitudinal factors are all significant. The positive attitudes to globalisation and job security in

Asia are consistent with national policy frameworks, an elite consensus and relatively weak

and disorganised labour movements. The one country with a reasonably strong and organised

union movement, South Korea, is also the country with the highest level of respondents

linking job insecurity and globalisation. In Europe, by contrast the variation in country results

suggests national factors of other kinds may be pertinent. Sweden and Ireland are the

countries with the highest level of engagement in international trade but with the lowest

incidence of concern about job insecurity. As noted earlier, (and following Esping-

Anderson’s categorisation), Sweden represents the only ‘social democratic’ welfare state in

this present sample. For its part, Ireland is closest to the development states of Asia in its

national economic strategy (O’Hearn, 1998).

This analysis also sheds light on the constellation of attitudes associated with

globalisation and employment. The first point to note is very large number of citizens who


continue to hold their national governments responsible for employment outcomes. In

Europe, more than half the citizens in six of the nine countries covered consider employment

to be a national responsibility. The results in the other three countries (France, Germany,

Portugal) were just below 50 percent and may reflect attachment to European institutions.

The results were even stronger in Asia with only one country (the Philippines) recording

results lower than 50 percent. The political significance of this outcome in relation to

unemployment is enhanced by the fact that pluralities in most of these same countries support

‘all-country’ action on other issues such as human rights and the environment.

Narrower constituencies link positive and negative employment outcomes to globalisation.

In Europe, the negative cohort is largest in the southern states and the UK (unweighted

average 30 percent). In the conservative welfare states of Germany and France this cohort

equals 24 percent, in Sweden 11 percent and in Ireland 13 percent. In Asia the unweighted

average of negative respondents is much lower at 18 percent, with South Korea recording the

highest number at 31 percent. This suggests a very different politics of employment and

globalisation in the two regions.

In Europe, there is some evidence for Rodrik’s speculation that globalisation is producing

a new and substantial cleavage between winners and losers. Potential hostile constituencies

number around 25 percent or more of national electorates in all countries save for Sweden

and Ireland. What circumstances and attitudes are especially salient with this cohort? A

variety of factors covering respondent’s economic situation, values, attitudes to national

governments, attitudes to national and supra-national institutions were tested along with

‘objective’ data on international economic engagement. The regression model sought to

identify inter alia the specific factors differentiating positive from negative responses.

Negative respondents were especially likely to feel economically insecure and particularly

worried about immigration, to conceive globalisation as Americanisation, and to discount

their government’s handling of unemployment. They were also more likely to see their

national governments as having an effect on daily life. Unlike analogous studies, support for

restriction of imports did not correlate with negative views about the impact of globalisation

on job security. Nor did positions on a conventional left-right scale.12

Those who see globalisation as a negative influence on job security are apparently open to

mobilisation from one of two directions. They can be diverted into a politics of blame – with

12 Jean Blondel's forthcoming analysis of the ASES data has also noted that the left-right 'divide has little explanatory power with respect to policy preferences; citizens do not appear to relate to economic and social issues by means of the Left-Right dimension.'


America and immigrants as the two targets suggested by the above findings; or they can be

reconciled to economic globalisation through national policies that ease insecurity and

facilitate adjustment and adaptation. As noted earlier in discussion of the Swank and Betz

findings, the ‘politics of blame’ is the hallmark of the right wing populist parties. There

would also seem to be potential for these parties to extend their electoral base if populist right

parties can link their immediate agendas to broader public sentiments associating national

governments with employment outcomes.

Alternative strategies for easing fears about job insecurity and economic globalisation are

suggested in the differentiated but proactive approaches of Ireland and Sweden. In Sweden’s

case, this involves a high employment outcome based on a state sponsored service sector. In

Ireland’s case, it involves a ‘developmental’ approach to economic governance backed by a

strong effort to build a political consensus around longer-term economic and social strategies.

Scharpf’s reservations about the difficulties of imitating Swedish outcomes in the

‘conservative/continental’ welfare states relate to the transfer payment structure of their

welfare systems, which inhibit employment creation in the public sector, and strong

resistance to deregulation, which inhibits an American style growth in private services (1997,

p. 3). In Asia, by contrast, the politics of employment and globalisation is very differently

constituted. The first point to note is the very recent development of democratic political

institutions in almost all countries other than Japan (where it has substantial longevity) and

China (where non-democratic governance prevails). Democracy was introduced in Korea and

Taiwan early in the past decade. It was secured in Thailand only after bloody public

resistance to a military backed regime in 1992. It was restored in the Philippines in the late

1980s and in Indonesia in 1998. Singapore is only nominally democratic: it is a one party

state, which suppresses opposition, and Malaysia is a semi-democracy. ‘Strong’ party

systems that resemble, at least in some respects, western forms are present only in Taiwan

and Japan.13 Thailand has only one membership-based party and Malaysia has membership

parties but in a very constrained system. There is no strong ideological cleavage between the

rival protagonists for political power in any state and no strong cleavages over economic

strategies. As noted earlier, trade unions have no or only weak presence in all countries

except for Korea and Japan. Further, public attitudes have been formed in a context in which

the political leadership has continually emphasised the desirability of, and the gains from,

participation in global trade. Zaller (1992) has established in a different context the

13 On the character of political parties in the region see Blondel (2003, forthcoming)


importance of elite consensus in shaping opinion, a finding that may be relevant here.

Moreover, and largely in spite of the financial crisis of 1997, outcomes have more or less

matched these promises. National populations have experienced real material benefits as the

general rise in per capita incomes and, until recently, falling income inequality attests. In

addition, participation in education has been extended in all countries although some, such as

Thailand and Indonesia, have a long way to go to match the performance of the East Asian

states (Korea and Taiwan). Finally, the American market has been a primary destination for

regional exports and the rapid recovery from the 1997 crisis in regional states was to no small

degree aided by the continued high demand for electronic and other goods by US businesses

and consumers.

This is the context in which the Asian results can be interpreted. Except for Japan, and

possibly Korea, it would appear that region-wide national commitments to export-led

economic growth and the absence of any significant opposition to this strategy amongst

political elites has framed the development of public attitudes that are very positive about the

benefits of globalisation for employment. This is a result not unlike that realised in Ireland.

Young people in particular hold positive views and this doubtless reflects generational

engagement with and attachment to ‘modernisation’ in all its aspects. Japan may be a special

case, as we noted earlier, with a much lower level of engagement in, and acknowledgement

of, globalisation issues. In Korea, there are some signs of conflict over economic strategy and

institutions, partly stemming from the growing influence of militant labour and their nascent

political representatives (JoongAng Iibo, 2002).

The one finding in the survey results that merits especial comment concerns the link

between schooling and attitudes to employment and globalisation. The results on schooling

are a more reliable indicator of skill levels in Asia than in Europe because of the greater

importance of school education in overall skill development. Those with less schooling in the

Asian states are more likely to identify a positive relationship between globalisation and

employment. This result is contrary to the findings of other studies (Mayda and Rodrik;

Scheve and Slaughter, cited earlier) where respondents with lower skill levels were much

more likely to be hostile to a related issue of free trade. In Asia, this may point to an

association between education and more critical political views. It may also point to the

particular impacts of the financial crisis on the nascent middle class in particular countries.

Both these speculations require further testing. If the former holds, this may point to the

progressive development of a more critical national politics – but this remains a distant

prospect.


In sum, the politics of globalisation and employment follows quite different trajectories in

Europe and Asia. In the former case, there is strong evidence of a new cleavage made up of

globalisation ‘losers’ that, in most countries, present policy frameworks have been unable to

contain or reconcile. In the absence of remedial policy frameworks, this constituency is

sufficiently large to make a significant impact on national politics. This impact could be

direct in those states where the political opportunity structure favours minority representation

or indirect in states with other than proportional voting systems: here, major parties,

particularly conservative parties, are recognising the political benefits accruing from

globalisation fears, particularly on the troubling issue of immigration and asylum seekers. In

Asia, by contrast, consensus at the level of the political leadership (matched by more or less

consistent material outcomes) in all states would appear to have shaped very positive views

amongst their publics. So long as these conditions persist, a favourable balance of public

opinion might be expected to be sustained.


APPENDIX A. Definitions of Variables The following provides information about the variables and how the data from the ASES Survey was coded. JOBSEC - Response to ‘it is said we now live in an age when all sorts of things (for example, products, money, people and information) move around the world much more than they used to. Please tell me whether this kind of movement has any effect on your own life [in your job security] and whether the overall effect has been a good thing or a bad thing.’ [Categorical: has good effect, has bad effect, has effect but is neither good nor bad, has no effect, don't know] AGE - Age in years GENDER - [female (1), male (0)] SCHOOL - Number of years of school education excluding apprenticeships GLOBALJOB - Respondent's job involves contact with organisations or people in other countries [applies (1), does not apply (0)] WORKINS - Scaled response about the level of respondent worry about ‘work situation’ [very worried (3), somewhat worried (2), not at all worried (1), don’t know] LIVSTAN - Scaled response to ‘how would you describe your household’s living standards’ [high (5), relatively high (4), average (3), relatively low (2), low (1)] LEFTRT - Scaled response to ‘how would you place your views on the [left-right] scale?’ [extreme left (1) to extreme right (10), don’t know] COMPGOOD - Scaled response to ‘competition is good because it stimulates people to develop new ideas’ [strongly agree (5), agree (4), neither agree nor disagree (3), disagree (2), strongly disagree (1), don't know] NATPRIDE - Respondent’s pride in being [their nationality] [very proud (4), somewhat proud (3), not so proud (2), not proud at all (1), don't know] AMERICA - Scaled response to ‘some people say that all this international influence is really a form of Americanisation. Do you think this is a fair description?’ [Fair (1), Not fair and don't know combined (0)] WORRYEC - Scaled response to ‘when thinking specifically about the situation in [country], how worried are you about the economy?’ [very worried (3), somewhat worried (2), not at all worried (1), don’t know] WORRYUE - Scaled response to ‘when thinking specifically about the situation in [country], how worried are you about unemployment?’ [very worried (3), somewhat worried (2), not at all worried (1), don’t know] WORRYIM - Scaled response to ‘when thinking specifically about the situation in [country], how worried are you about immigration?’ [very worried (3), somewhat worried (2), not at all worried (1), don’t know] GOVEC - Scaled response to ‘how well to you think [national] government is dealing with the economy?’ [very well (4), fairly well (3), not so well (2), not well at all (1), don't know] GOVUE - Scaled response to ‘how well to you think [national] government is dealing with unemployment?’ [very well (4), fairly well (3), not so well (2), not well at all (1), don't know] GOVIM - Scaled response to ‘how well to you think [national] government is dealing with immigration?’ [very well (4), fairly well (3), not so well (2), not well at all (1), don't know]


GOVIM2 - Scaled response to ‘how well to you think [national] government is dealing with immigration?’ [very well or fairly well combined (1), not so well, not well at all and don't know combined (0)] CONFPOLP - Scaled response to ‘could you tell me how much confidence you have in the political parties in [country]?’ [a great deal (4), quite a lot (3), not much (2), none at all (1), don't know, haven't thought much about it] CONFPOLL - Scaled response to ‘could you tell me how much confidence you have in the main political leaders in [country]?’ [a great deal (4), quite a lot (3), not much (2), none at all (1), don't know, haven't thought much about it] CONFBBUS - Scaled response to ‘could you tell me how much confidence you have in big business in [country]?’ [a great deal (4), quite a lot (3), not much (2), none at all (1), don't know, haven't thought much about it] GOVRES - Scaled response to ‘the government should take responsibility for ensuring everyone either has a job or is provided with adequate social welfare’ [strongly agree (5), agree (4), neither agree nor disagree (3), disagree (2), strongly disagree (1), don't know] LIMIMP - Scaled response to ‘[your country] should limit the imports of foreign products’ [strongly agree(5), agree (4), neither agree nor disagree (3), disagree (2), strongly disagree (1), don't know] EFFECTGOV - Scaled response to ‘how much effect [do] you think the activities, decisions and so on of the [national] government have on your day-to-day life?’ [great effect (3), some effect (2), no effect and don't know combined (1)] EFFECTEU - Scaled response to ‘how much effect [do] you think the activities, decisions and so on of the European Union have on your day-to-day life?’ [great effect (3), some effect (2), no effect and don't know combined (1)] EFFECTAS - Scaled response to ‘how much effect [do] you think the activities, decisions and so on of ASEAN have on your day-to-day life?’ [great effect (3), some effect (2), no effect and don't know combined (1)] EFFECTUN - Scaled response to ‘how much effect [do] you think the activities, decisions and so on of the UN and its various agencies have on your day-to-day life?’ [great effect (3), some effect (2), no effect and don't know combined (1)] EFFECTMN - Scaled response to ‘how much effect [do] you think activities, decisions and so on of multinational companies have on your day-to-day life?’ [great effect (3), some effect (2), no effect and don't know combined (1)] UNEMP - National unemployment rates 2000 from World Bank Online. Taiwan figure estimate for 2000 from CIA World Factbook. EXPGDP - National export/GDP ratios 1999 from World Bank Online. Taiwan estimate for 2000 calculated from CIA World Factbook. LTGR - Long-term growth rates in GDP 1990-2000 from World Bank database online Taiwan estimate for 2000 from CIA World Factbook. FORPOP - Percentage of foreign-born born workers in the labour force 1999 from OECD in Figures 2001 edition. NETMIG - Five-year average national net migration rate 1995-2000 from World Bank Online. NB: ‘don't know’ responses were excluded from the scaled variables unless recoding information specifies.


APPENDIX B1: Europe model . mlogit jobsec age gender livstan workins globaljob natpride /* > */ compgood america worryec worryue worryim govue govres /* > */ effectgov effecteu effectmn unemp expgdp ltgr forpop /* > */ netmig if region==2, b(1) Iteration 0: log likelihood = -11624.772 Iteration 1: log likelihood = -10856.092 Iteration 2: log likelihood = -10836.08 Iteration 3: log likelihood = -10835.948 Iteration 4: log likelihood = -10835.948 Multinomial regression Number of obs = 7624 LR chi2(84) = 1577.65 Prob > chi2 = 0.0000 Log likelihood = -10835.948 Pseudo R2 = 0.0679 ------------------------------------------------------------------------------ jobsec | Coef. Std. Err. z P>|z| [95percent Conf. Interval] -------------+---------------------------------------------------------------- has bad ef~t | age | .003694 .0023343 1.58 0.114 -.0008811 .0082691 gender | -.0352765 .0691157 -0.51 0.610 -.1707409 .1001878 livstan | -.2344073 .0512809 -4.57 0.000 -.334916 -.1338986 workins | .301628 .0492601 6.12 0.000 .2050801 .398176 globaljob | -.1690186 .0878401 -1.92 0.054 -.3411821 .0031449 natpride | -.2095237 .0463826 -4.52 0.000 -.3004318 -.1186155 compgood | -.1536687 .0459113 -3.35 0.001 -.2436531 -.0636843 america | .3116846 .0696042 4.48 0.000 .1752629 .4481063 worryec | .1198036 .0598245 2.00 0.045 .0025496 .2370575 worryue | -.0702896 .0653546 -1.08 0.282 -.1983823 .0578032 worryim | .2173567 .0484566 4.49 0.000 .1223835 .3123298 govue | -.3628764 .0475348 -7.63 0.000 -.456043 -.2697098 govres | -.0985861 .0443521 -2.22 0.026 -.1855146 -.0116575 effectgov | .2470398 .0612404 4.03 0.000 .1270108 .3670687 effecteu | -.2127763 .0650101 -3.27 0.001 -.3401938 -.0853588 effectmn | -.0085834 .0537759 -0.16 0.873 -.1139821 .0968153 unemp | -.0164727 .012649 -1.30 0.193 -.0412643 .0083188 expgdp | -.0128587 .006366 -2.02 0.043 -.0253358 -.0003815 ltgr | .1197793 .0831065 1.44 0.150 -.0431065 .282665 forpop | -.0187225 .0226925 -0.83 0.409 -.0631989 .0257539 netmig | -.1464483 .0421315 -3.48 0.001 -.2290245 -.0638722 _cons | 2.345587 .4420626 5.31 0.000 1.47916 3.212013 -------------+---------------------------------------------------------------- has effect~n | age | .0027948 .0022441 1.25 0.213 -.0016036 .0071932 gender | -.0000123 .0671977 -0.00 1.000 -.1317173 .1316927 livstan | .0079781 .0499231 0.16 0.873 -.0898694 .1058255 workins | .0068174 .0498343 0.14 0.891 -.0908561 .1044909 globaljob | -.1801041 .0835135 -2.16 0.031 -.3437876 -.0164207 natpride | -.1226257 .0456853 -2.68 0.007 -.2121671 -.0330842 compgood | -.1867938 .044955 -4.16 0.000 -.274904 -.0986837 america | .0196064 .0670356 0.29 0.770 -.111781 .1509937 worryec | -.0815629 .0582363 -1.40 0.161 -.195704 .0325782 worryue | -.1519344 .0607743 -2.50 0.012 -.2710499 -.0328188 worryim | -.052252 .0467274 -1.12 0.263 -.1438361 .0393321


govue | -.1764163 .0462843 -3.81 0.000 -.2671319 -.0857008 govres | -.1637217 .0412836 -3.97 0.000 -.2446361 -.0828072 effectgov | .0367772 .0605592 0.61 0.544 -.0819166 .155471 effecteu | -.0953481 .0641595 -1.49 0.137 -.2210984 .0304022 effectmn | -.0452208 .052827 -0.86 0.392 -.1487598 .0583183 unemp | .0366034 .0124783 2.93 0.003 .0121465 .0610603 expgdp | .0053095 .0059529 0.89 0.372 -.0063579 .016977 ltgr | -.0285368 .0779137 -0.37 0.714 -.1812448 .1241711 forpop | .0078379 .0217997 0.36 0.719 -.0348887 .0505645 netmig | -.2383601 .0427757 -5.57 0.000 -.3221988 -.1545213 _cons | 3.010016 .4249821 7.08 0.000 2.177066 3.842965 -------------+---------------------------------------------------------------- has no eff~t | age | .0208673 .0023285 8.96 0.000 .0163036 .025431 gender | -.0085589 .0714943 -0.12 0.905 -.1486853 .1315674 livstan | -.1576837 .052647 -3.00 0.003 -.2608698 -.0544976 workins | -.2354779 .0565869 -4.16 0.000 -.3463861 -.1245696 globaljob | -.2837074 .0934841 -3.03 0.002 -.4669329 -.100482 natpride | -.0132283 .0500483 -0.26 0.792 -.1113211 .0848645 compgood | -.1042796 .0494391 -2.11 0.035 -.2011785 -.0073807 america | .0386454 .0713955 0.54 0.588 -.1012871 .1785779 worryec | -.2767013 .0621929 -4.45 0.000 -.3985971 -.1548055 worryue | -.1403667 .0636771 -2.20 0.027 -.2651716 -.0155619 worryim | -.0674208 .0496583 -1.36 0.175 -.1647493 .0299077 govue | -.262111 .0501782 -5.22 0.000 -.3604583 -.1637636 govres | -.1836094 .0437799 -4.19 0.000 -.2694164 -.0978024 effectgov | -.1265195 .0642676 -1.97 0.049 -.2524817 -.0005573 effecteu | -.1789688 .0687066 -2.60 0.009 -.3136312 -.0443063 effectmn | -.2298951 .0580016 -3.96 0.000 -.3435761 -.1162142 unemp | -.0182824 .0133238 -1.37 0.170 -.0443966 .0078318 expgdp | .0016482 .0061722 0.27 0.789 -.0104491 .0137456 ltgr | .2446267 .0833738 2.93 0.003 .0812171 .4080364 forpop | .0504172 .0244531 2.06 0.039 .0024901 .0983444 netmig | -.4271757 .0511156 -8.36 0.000 -.5273604 -.3269911 _cons | 3.695704 .4500775 8.21 0.000 2.813569 4.57784 -------------+---------------------------------------------------------------- don't know | age | .0247328 .0035469 6.97 0.000 .0177811 .0316846 gender | .4080571 .1106679 3.69 0.000 .1911519 .6249623 livstan | -.1146884 .0815367 -1.41 0.160 -.2744975 .0451206 workins | .1270137 .080323 1.58 0.114 -.0304165 .2844439 globaljob | -.6917388 .172566 -4.01 0.000 -1.029962 -.3535156 natpride | -.124319 .0746988 -1.66 0.096 -.2707259 .0220878 compgood | -.1722468 .0715511 -2.41 0.016 -.3124844 -.0320092 america | -.1701774 .1095024 -1.55 0.120 -.3847982 .0444434 worryec | .0225837 .0962397 0.23 0.814 -.1660426 .21121 worryue | -.2263977 .1016007 -2.23 0.026 -.4255314 -.0272641 worryim | .0873237 .078072 1.12 0.263 -.0656947 .240342 govue | -.2003148 .0753446 -2.66 0.008 -.3479875 -.0526422 govres | -.0622555 .069634 -0.89 0.371 -.1987356 .0742246 effectgov | .0878116 .0955971 0.92 0.358 -.0995554 .2751785 effecteu | -.3523377 .1056848 -3.33 0.001 -.5594762 -.1451993 effectmn | -.1835899 .0905584 -2.03 0.043 -.3610811 -.0060987 unemp | .0811336 .0203069 4.00 0.000 .0413327 .1209344 expgdp | .0032441 .0102708 0.32 0.752 -.0168864 .0233745


ltgr | .1151153 .1297822 0.89 0.375 -.1392532 .3694837 forpop | -.0032152 .0369355 -0.09 0.931 -.0756074 .0691771 netmig | -.1682978 .0664447 -2.53 0.011 -.2985271 -.0380685 _cons | -.2352775 .6894862 -0.34 0.733 -1.586646 1.116091 ------------------------------------------------------------------------------ (Outcome jobsec==has good effect is the comparison group)


APPENDIX B2: Asia Model . mlogit jobsec age gender school workins natpride /* > */ compgood worryue govec govim2 confpoll effectgov effectas effectmn unemp expgdp ltgr net > mig if region==1, b(1) Iteration 0: log likelihood = -9806.6067 Iteration 1: log likelihood = -9305.1514 Iteration 2: log likelihood = -9291.6812 Iteration 3: log likelihood = -9291.5551 Iteration 4: log likelihood = -9291.555 Multinomial regression Number of obs = 6719 LR chi2(68) = 1030.10 Prob > chi2 = 0.0000 Log likelihood = -9291.555 Pseudo R2 = 0.0525 ------------------------------------------------------------------------------ jobsec | Coef. Std. Err. z P>|z| [95percent Conf. Interval] -------------+---------------------------------------------------------------- has bad ef~t | age | .0089207 .0029892 2.98 0.003 .0030619 .0147795 gender | -.156248 .0758605 -2.06 0.039 -.3049318 -.0075642 school | .0325626 .0107164 3.04 0.002 .0115588 .0535663 workins | .4601771 .0523159 8.80 0.000 .3576398 .5627144 natpride | -.1314244 .0589314 -2.23 0.026 -.2469278 -.015921 compgood | -.1431021 .0510058 -2.81 0.005 -.2430716 -.0431326 worryue | .0203406 .067992 0.30 0.765 -.1129212 .1536025 govec | -.1909438 .0545032 -3.50 0.000 -.297768 -.0841195 govim2 | -.4307003 .0817703 -5.27 0.000 -.5909672 -.2704333 confpoll | -.1440817 .0470132 -3.06 0.002 -.2362259 -.0519376 effectgov | -.1008931 .0584992 -1.72 0.085 -.2155495 .0137633 effectas | -.0640148 .0694288 -0.92 0.357 -.2000927 .0720631 effectmn | -.0946204 .0684729 -1.38 0.167 -.2288247 .0395839 unemp | -.0439505 .0192118 -2.29 0.022 -.0816049 -.0062961 expgdp | -.0045301 .0020651 -2.19 0.028 -.0085776 -.0004826 ltgr | -.0127088 .0382963 -0.33 0.740 -.0877682 .0623507 netmig | .0585587 .0105 5.58 0.000 .037979 .0791383 _cons | .7464424 .4702932 1.59 0.112 -.1753153 1.6682 -------------+---------------------------------------------------------------- has effect~n | age | .006381 .0025461 2.51 0.012 .0013908 .0113712 gender | .0699511 .0643418 1.09 0.277 -.0561566 .1960588 school | .0415777 .0091818 4.53 0.000 .0235817 .0595736 workins | .1532277 .0447825 3.42 0.001 .0654556 .2409998 natpride | -.1560852 .0519927 -3.00 0.003 -.257989 -.0541814 compgood | -.2043875 .0431357 -4.74 0.000 -.2889319 -.119843 worryue | -.1710627 .0559473 -3.06 0.002 -.2807175 -.0614079 govec | -.1514461 .0467288 -3.24 0.001 -.2430329 -.0598594 govim2 | -.433157 .0700098 -6.19 0.000 -.5703736 -.2959404 confpoll | -.0911899 .0408247 -2.23 0.026 -.1712048 -.011175 effectgov | .0064846 .05019 0.13 0.897 -.091886 .1048552 effectas | -.1886535 .0596608 -3.16 0.002 -.3055865 -.0717205 effectmn | -.1570751 .0585887 -2.68 0.007 -.2719068 -.0422433 unemp | .0481636 .0158608 3.04 0.002 .0170771 .0792501


expgdp | -.0005591 .0018112 -0.31 0.758 -.0041091 .0029908 ltgr | -.1438131 .0344758 -4.17 0.000 -.2113844 -.0762418 netmig | .0535853 .00867 6.18 0.000 .0365924 .0705782 _cons | 2.434234 .400667 6.08 0.000 1.648941 3.219527 -------------+---------------------------------------------------------------- has no eff~t | age | .0152046 .0030314 5.02 0.000 .0092631 .0211461 gender | .0220499 .0796403 0.28 0.782 -.1340423 .1781421 school | .0459091 .0109848 4.18 0.000 .0243792 .067439 workins | -.150064 .0556739 -2.70 0.007 -.2591829 -.0409451 natpride | -.1534155 .062276 -2.46 0.014 -.2754742 -.0313569 compgood | -.2050161 .052389 -3.91 0.000 -.3076966 -.1023356 worryue | -.2063818 .0671389 -3.07 0.002 -.3379717 -.074792 govec | -.1517176 .0575419 -2.64 0.008 -.2644976 -.0389376 govim2 | -.4623872 .0869602 -5.32 0.000 -.6328261 -.2919482 confpoll | -.0129419 .0505481 -0.26 0.798 -.1120144 .0861305 effectgov | -.1931256 .0605362 -3.19 0.001 -.3117744 -.0744768 effectas | -.2212912 .0775546 -2.85 0.004 -.3732954 -.0692871 effectmn | -.2660229 .0761561 -3.49 0.000 -.4152862 -.1167596 unemp | -.0592921 .0200912 -2.95 0.003 -.0986702 -.019914 expgdp | .0016378 .0021995 0.74 0.456 -.0026731 .0059487 ltgr | -.2491492 .0399952 -6.23 0.000 -.3275384 -.17076 netmig | .0286967 .0107598 2.67 0.008 .0076078 .0497856 _cons | 3.437252 .4776871 7.20 0.000 2.501003 4.373502 -------------+---------------------------------------------------------------- don't know | age | .0248076 .0046452 5.34 0.000 .0157032 .0339121 gender | .341144 .1241466 2.75 0.006 .0978212 .5844669 school | .0657666 .0162736 4.04 0.000 .0338709 .0976624 workins | .0935609 .0854298 1.10 0.273 -.0738784 .2610001 natpride | -.2797112 .0884291 -3.16 0.002 -.4530291 -.1063933 compgood | -.2188661 .07898 -2.77 0.006 -.3736641 -.064068 worryue | -.1587327 .1070474 -1.48 0.138 -.3685418 .0510764 govec | -.1704054 .0900175 -1.89 0.058 -.3468363 .0060256 govim2 | -.7206288 .1419959 -5.07 0.000 -.9989356 -.4423219 confpoll | .0179056 .0791988 0.23 0.821 -.1373212 .1731324 effectgov | .0117668 .0897903 0.13 0.896 -.1642188 .1877525 effectas | -.6153287 .132803 -4.63 0.000 -.8756177 -.3550397 effectmn | -.4003507 .1283908 -3.12 0.002 -.651992 -.1487094 unemp | -.092213 .0363323 -2.54 0.011 -.163423 -.0210031 expgdp | -.0102832 .0036561 -2.81 0.005 -.017449 -.0031173 ltgr | -.0684288 .0556237 -1.23 0.219 -.1774491 .0405916 netmig | .0631554 .0203419 3.10 0.002 .023286 .1030248 _cons | 1.72956 .7203364 2.40 0.016 .317727 3.141394 ------------------------------------------------------------------------------ (Outcome jobsec==has good effect is the comparison group)


Bibliography Ali, T. 2002, Clash of Fundamentalisms: Crusades, Jihads and Modernity, Verso, New York. Bhagwati, J. 2002, Free Trade Today, Princeton University Press, Princeton, NJ. Blondel, J. 2003, Parties and Party Systems in East and South East Asia in Marsh, I. ed., Parties, Leadership and Governance in East and South East Asia: A Comparative Study, forthcoming. Bronfenbrenner, K. 1997, ‘Class War in the USA’, Multinational Monitor, vol. 18, no. 3, pp. 1-5. Campos, E. and H. Root, 1996. The Key to the Asian Miracle: Making Shared Growth Credible, The Brookings Institution, Washington. Chomsky, N. 2001, September 11, Seven Stories Press. Economist Intelligence Unit, 2002, World Investment Prospects, London. Esping-Andersen, G. 1990, The Three Worlds of Welfare Capitalism, Princeton University Press, Princeton, NJ. Fligstein, N. and F. Merand, 2001, Globalisation or Europeanisation? Evidence of the European Economy Since 1980, Paper prepared for the conference ‘Shareholder Value-Capitalism and Globalisation, Hamburg, 10-12 May. Fligstein, N. 1997, Is Globalisation the Cause of the Crises of Welfare States? Paper prepared for the annual meeting of the American Sociological Association, Toronto, Canada. Fraser, J and J. Oppenheim, What’s New About Globalisation? The McKinsey Quarterly, No. 2, 1997, pp.169-179. Friedman, T. 2000, The Lexus and the Olive Tree, Anchor Books Edition, New York. Giddens, A. 1990, The Consequences of Modernity, Polity Press, Cambridge. Hirst, P. and D. Held 2002, Globalisation after 11 September, openDemocracy dialogues, www.opendemocracy.com Hirst, P. and G. Thompson, Globalisation in Question, Polity Press, Cambridge. Hook, G. 1996, Japan and the Construction of Asia-Pacific, in Andrew Gamble and Anthony Payne (eds), Regionalism and World Order, Macmillan Press, Basingstoke. Hout, M. 2002, ‘Growing Inequality and Social Mobility’, paper delivered to Symposium on Inequality and Exclusion, XV World Congress of Sociology, Brisbane, Australia, July. Irwin, D. A. 2002, Free Trade Under Fire, Princeton University Press, Princeton. NJ.

http://www.opendemocracy.com/


JoongAng Iibo 2002, 'Progressive eyes Blue House', June 29, online edition. Kapstein, E. 2000, Winners and Losers in the Global Economy, International Organisation, Vol. 54, No. 2, pp. 359-384. Kim, H-K., M. Muramatsu, T. J. Pempel and K. Yamamura eds., The Japanese Civil Service and Economic Development, Catalysts of Change, Clarendon Press, Oxford. Kohli A., C-I. Moon and G. Sorenson 2002, States, Markets and Just Growth, United Nations University Press, Tokyo. Krugman, P. 1996, First Do No Harm, Foreign Affairs, July-August, 1996, pp.164-170. Manski C. and J. D. Straub, 1995, ‘Worker Perceptions of Job Insecurity in the mid-1990s: Evidence from the Survey of Economic Expectations’, Working Paper 6908, NBER, Cambridge, Mass., pp. 1-48) Meidner, R. 1993, Why Did the Swedish Model Fail?, Miliband, R and L. Panitch, eds., The Socialist Register. Mayda A. and D. Rodrik 2002, Why are Some People (and Countries) More Protectionist than Others? Unpublished paper. OECD 1997, Employment Outlook, OECD, Paris. OECD, 2001 International Survey Research, cited Trade Union Congress (UK), May 2002, Globalisation: Myths and Realities. O’Hearn, D. 1998, Inside the Celtic Tiger, Pluto Press, London Pierson, P.1994, Dismantling the Welfare State? Cambridge University Press, New York. Ramesh, M. 2000, Welfare Capitalism in Southeast Asia, Macmillan, Basingstoke. Rydgren, J. 2002, Radical Right Populism in Sweden: Still a Failure, But for How Long?, Scandinavian Political Studies, Vol. 25, No. 1, pp.27-56. Rose, R. 2002, Are People for Open Trade, Protectionist, or Just Confused?, www.opendemocracy.net, pp. 1-4 Rowthorn, R.E. and R. Ramaswamy 1998, ‘Growth, Trade and De-industrialisation’, IMF Working Paper WP/98/60 Scharpf, F. 1997, European Distinctions, Boston Review, Summer, pp. 1-4. Scheve K.and M. Slaughter, 2001, What determines individual trade policy preferences? Journal of International Economics, Vol. 54, pp. 267-292. Scott, Bruce R. 2001, The Great Divide in the Global Village, Foreign Affairs, January/February

http://www.opendemocracy.net/


Stephens, J. D. 2000, Is Swedish Corporatism Dead? Thoughts on its Supposed Demise in the Light of the Abortive ‘Alliance for Growth’ in 1998. Paper presented to 12th International Conference of Europeanists, Council of European Studies, Chicago, 30 March - 1 April. Sum, Ngai-Lim 1996, The NICs and Competing Strategies of East Asian Regionalism, in Andrew Gamble and Anthony Payne (eds), Regionalism and World Order, Macmillan Press, Basingstoke . Svallfors, S 1997, Worlds of Welfare and Attitudes to Redistribution: A Comparison of Eight Western Nations. European Sociological Review, No 3, 1997, pp. 283-304 Swank, D and H-G Betz, 2002, Globalisation, the Welfare State and Right-Wing Populism in Western Europe, Socio-Economic Review, forthcoming. Swank, D. 2002, Withering Welfare? Globalisation, Political Economic Institutions and Contemporary Welfare States, in L.Weiss (ed), States in the Global Economy: Bringing Domestic Institutions Back In, Cambridge University Press, Cambridge. Taylor, T. 2002, The Truth about Globalisation, The Public Interest, Spring 2002, pp.24-44 Therborn, G. 2002, ‘Dimensions and Processes of Global Inequalities’, paper delivered to Symposium on Inequality and Exclusion, XV World Congress of Sociology, Brisbane, Australia, July. Wade, R. 2000, Wheels Within Wheels: Rethinking the Asian Crisis and the Asian Model, Annual Review of Political Science, Palo Alto. Ca. Wilensky, H. 2002, Rich Democracies: Political Economy, Public Policy and Performance, University of California Press, Berkeley, Ca. Woo-Cummings, M. 1999, The Developmental State, Cornell University Press, Ithaca, NY. Zaller, J. 1992, The Nature and Origins of Mass Opinion, Cambridge University Press, Cambridge.

Job Insecurity and Globalisation: What Factors Shape ... · Job insecurity in public discussion refers to a number of inter-related situations: actual unemployment, the expectation

Documents