Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The influence of sociotropic judgment on public perceptions of trade liberalization : distributional… Lee, Hyunji 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_fall_lee_hyunji.pdf [ 1.37MB ]
Metadata
JSON: 24-1.0072187.json
JSON-LD: 24-1.0072187-ld.json
RDF/XML (Pretty): 24-1.0072187-rdf.xml
RDF/JSON: 24-1.0072187-rdf.json
Turtle: 24-1.0072187-turtle.txt
N-Triples: 24-1.0072187-rdf-ntriples.txt
Original Record: 24-1.0072187-source.json
Full Text
24-1.0072187-fulltext.txt
Citation
24-1.0072187.ris

Full Text

 THE INFLUENCE OF SOCIOTROPIC JUDGMENT ON PUBLIC PERCEPTIONS OF TRADE LIBERALIZATION: DISTRIBUTIONAL JUDGMENT AND INSTITUTIONAL FACTORS    by    Hyunji Lee      A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF    DOCTOR OF PHILOSOPHY  in   The Faculty of Graduate Studies    (Political Science)    THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)    August 2011     © Hyunji Lee, 2011   ii ABSTRACT  This dissertation sheds new light on the question of why some individuals and some countries are more protectionist than others, by applying findings from public opinion studies to the topic.  More specifically, by addressing the gap between the predictions of economic trade theories and the findings of public opinion studies, I examine non-economic determinants of individual attitudes toward trade policies, and institutional factors as an intervening variable that could mediate or exacerbate protectionist sentiment.   This dissertation consists of five separate papers each of which tackles a different theoretical or empirical puzzle: why is there a discrepancy in views on trade liberalization between economists and the public; why are females more protectionist than males; does more spending on welfare bring about more public support for openness; what explains the recent protectionist backlash in Korea, an export-oriented economy where there is a public consensus on the positive impact of trade on the national economy; and does democratization lead to more economic openness as predicted by factor endowment models?  Each chapter provides answers to each puzzle by utilizing three different methods that include survey experiments, survey analyses, and content analyses.  The findings of the five papers converge on the following two: (1) positive effects of income growth on support for trade are significantly offset by concerns with the effect of trade liberalization on domestic social and economic arrangements, e.g., increasing inequality and poverty; (2) the negative effects of such communitarian concerns on support for trade are magnified by a lack of public confidence in their government’s effectiveness and responsiveness.  The importance of the communitarian critiques of trade liberalization in shaping trade attitudes not only suggests that trade-induced economic growth does not necessarily lead to public support for greater trade openness; but it also suggests that the prediction of the factor endowment models that the electorates (the median voter) in capital scarce countries prefer more openness is flawed.  The findings also suggest that new democracies often characterized by weak political institutions and rule of law are not necessarily in a better position than their authoritarian counterparts to garner public support for trade liberalization.     iii TABLE OF CONTENTS   ABSTRACT.......................................................................................................................................... ii TABLE OF CONTENTS................................................................................................................. iii LIST OF TABLES............................................................................................................................... v LIST OF FIGURES...........................................................................................................................vi ACKNOWLEDGEMENTS ........................................................................................................... vii  INTRODUCTION ............................................................................................................................. 1  CHAPTER 1 DISTRIBUTIONAL JUDGMENT IN INDIVIDUAL PREFERENCES OVER TRADE LIBERALIZATION...........................................................................................18 1.1 Introduction ............................................................................................18 1.2 Externalities of trade and trade preferences ...............................................21 1.2.1 Positive externality: Growth .......................................................22 1.2.2 Negative externality: Inequality ...................................................23 1.2.3 Negative externality: Poverty ......................................................27 1.3 Method: Survey experiments ....................................................................29 1.4 Results: Effects of inequality and poverty on support for trade ..................32 1.5 Conclusion ..............................................................................................41  CHAPTER 2 THE SOURCE OF FEMALE PROTECTIONISM: THE CONSEQUENCE OF KNOWLEDGE GAP IN ECONOMICS?........................................44 2.1 Introduction ............................................................................................44 2.2 The gender gap in protectionist sentiments and sociotropic hypotheses......48 2.3 Explaining the gender gap in protectionist sentiment.................................55 2.3.1 Preview and hypotheses ..............................................................55 2.3.2 The survey experiment...............................................................58 2.3.3 Results: Economic literacy and the gender gap .................................61 2.3.4 Results: Economic man and social woman? ....................................67 2.4 Conclusions.............................................................................................70  CHAPTER 3 THE COMPROMISE OF EMBEDDED LIBERALISM AND GOVERNMENT CREDIBILITY.................................................................................................73 3.1 Introduction ............................................................................................73 3.2 Credible commitment and individual attitudes toward trade .......................76 3.3 Empirical test: Government credibility and protectionist sentiment............85 3.3.1 Description of the data and empirical approach................................85 3.3.2 Government credibility model ......................................................95 3.4 Conclusion ............................................................................................108    iv CHAPTER 4 TRUST IN GOVERNMENT AND PROTECTIONIST SENTIMENT: KOREAN PUBLIC OPINION ON THE KOREA-US FREE TRADE AGREEMENT (KORUS-FTA).................................................................................................................................110 4.1 Introduction: Framing the puzzle ...........................................................110 4.2 Political trust and protectionist sentiment................................................117 4.3 Decline in political trust in new democracies ...........................................123 4.4 Korean public opinion on KORUS FTA: The hypotheses .......................127 4.5 Data and method ...................................................................................135 4.6 Explaining trade policy preferences.........................................................140 4.7 Conclusion ............................................................................................144  CHAPTER 5 DOES DEMOCRACY ENCOURAGE TRADE LIBERALIZATION? FACTOR ENDOWMENTS AND PROTECTIONIST SENTIMENT IN LESS- DEVELOPED COUNTRIES......................................................................................................146 5.1 Introduction ..........................................................................................146 5.2 Theories and hypotheses ........................................................................150 5.2.1 Implications of public opinion studies for trade preferences.................150 5.2.2 The factor endowments model and trade preferences .........................156 5.3 Data and methods..................................................................................160 5.4 Results and discussion............................................................................167 5.5 Implications: Does democracy encourage trade openness? .......................175  CONCLUSION...............................................................................................................................177  REFERENCES................................................................................................................................183   v LIST OF TABLES   Table 1.1 Treatment effects on responses about import restrictions.........................................33 Table 1.2 Economic education and treatment effects ..................................................................34 Table 1.3 Job prospect and treatment effects ................................................................................36 Table 1.4 Logistic regression: Protectionist sentiment (Individual attitudes toward restrictions on foreign import) ........................................................................................39 Table 2.1: Survey questions related to trade in CCFR 2004 and gender differences ...............55 Table 2.2: Gender and attitudes toward agricultural subsidies ....................................................62 Table 2.3 The effect of economic literacy on the gender gap .....................................................62 Table 2.4 Individual support for agricultural subsidies – ideology, inequality-aversion, sympathy, and the gender gap (Full sample) .................................................................66 Table 2.5 Gender and the treatment effects...................................................................................68 Table 2.6: Logistic estimation: gender and treatment effects ......................................................69 Table 3.1 Summary data on individual attitudes toward trade (Asia-Europe Survey 2001)....16 Table 3.2 Government credibility model........................................................................................98 Table 4.3 Models of insecurity .......................................................................................................104 Table 5.1 Arguments against KORUS FTA and the U.S. beef (The Coding Scheme) .........139 Table 6.1 Country-level independent variables............................................................................166 Table 6.2 Multilevel logistic regression for protectionist sentiment: Factor endowment models...............................................................................................................................165 Table 6.3 Multilevel logistic regression for protectionist sentiment: Inequality and insecurity ...........................................................................................................................................173   vi LIST OF FIGURES   Figure 3.1 Scatterplot of protectionism and subjective evaluation of government effectiveness.......................................................................................................................91 Figure 3.2 Scatterplot of protectionism and subjective evaluation of welfare system .............93 Figure 3.3 Scatterplot of protectionism and perceived corruption.............................................93 Figure 3.4 Interaction effects between sociotropic insecurity and government effectivness ..... ...........................................................................................................................................106 Figure 4.1 Change in political trust (Confidence in parliament) ...............................................126 Figure 4.2 Changes in Korean public opinion on KORUS FTA during the negotiation .....129 Figure 4.3 Arguments against the FTA and the U.S. beef import ............................................141 Figure 5.1 Protectionist sentiment by country.............................................................................161  Figure 3.4 Interaction effects between sociotropic insecurity and government effectiveness   vii ACKNOWLEDGEMENTS  I would never have been able to finish my dissertation without the guidance of my mentors, help from friends and colleagues, and support from my family.  I am greatly indebted to my mentors at the University of British Columbia (UBC) – Maxwell Cameron, Paul Quirk, Angela O’Mahony, and Yves Tiberghien – for their support and guidance during my graduate years at UBC.  Max was an excellent advisor: He patiently put me back on track whenever I lost the forest for the trees. He helped me to stay focused and he gave me tremendous encouragement and support at crucial points of frustration during the whole process.  I am also deeply grateful to my other committee members, Paul Quirk and Angela O’Mahony.  Paul provided invaluable guidance and detailed advice in the process, from designing the project to developing it into a dissertation.  He was patient in pushing me to be more rigorous and to think clearly and logically.  The first two chapters of the dissertation would have been simply non-existent without his untiring help and guidance. I also benefited greatly from Angel’s professional advice and personal support.  Angel has been my role model: she was an inspirational mentor and a superb teacher.  I was blessed to have her in my committee and would like to thank her for unstinting advice and invaluable critiques of my work.  Finally, I want to express my deepest gratitude to Yves Tiberghien, who provided me with counsel, inspiration, and encouragement throughout the entire program.  In fact, Chapter 4 grew out of our collaborative research in Korea in 2008.  I thank him for his helpful and constructive comments on that chapter.  I have also benefited from comments and encouragement of my colleagues and friends in Canada.  Many thanks to Nathan Allen, Naranzul Bayasgalan, Juan Blanco, Na Guo, Kate Hecht, Junho Hwang, Tian Jie, Konrad Kalicki, Dongwoo Kim, Julie Kim, Jody Kliffer, Emtiyaz Khan, Hojun Lee, Joolee Lee, Myung-Shin Lee, Seonah Lee, Clare McGovern, Dale Marsden, Jeewon Min, Go Murakami, Heejin Park, Tommi Rebien, Miles Stauffer, Ho Suh, Mei Tan, Netina Tan, Doreen Tenerelli, Tashi Tsering, Erin Williams, and Yeohyun Yoon.  I would especially like to thank Erin, who graciously read the draft of the dissertation and offered astute substantive and editorial advice and valuable encouragement. I also owe special thanks to Netina for her warm, motherly support and professional advice throughout the process.  Emtiyaz and Konrad also deserve a particular mention.  Emti patiently answered my questions about data analysis, and directed me toward the SPIDA (Summer Program in Data Analysis) at York University.  And the often heated conversations I had with Konrad about welfare states and markets sparked new research ideas.  I also thank Amanda Bittner, Marc Andre Bodet, Jack Citrin, Vincent della Sala, and Laura Stephenson, for their comments on the chapters that I presented at conferences.  I am grateful to Kyung-Ae Park and Paul Evans at the Institute of Asian Research (IAR) for offering a pleasant work environment and support.  I also thank my officemates at the IAR for a stress-relieving and fun, but productive atmosphere.  I owe tremendous debt to my mentors, seniors, and friends in Korea.  I would not be where I am now without guidance of my master’s degree advisor, Byung-Kook Kim (BKK).  BKK is the very first person who taught me the joy of puzzle-solving political science research: I still remember his graduate seminar on comparative politics at Korea   viii University (KU) as an eye-opening learning experience for me.   He is also the one who guided me towards making the decision to study abroad; and his advice as a mentor enabled me to ground my research ideas in my homeland, Korea, while pushing me to place them in a broader comparative and theoretical context.  I also thank Jang Jip Choi, Young Jong Choi, Jung-In Kang and Nae-Young Lee for teaching me the fundamentals of political science. My seniors from KU, Jin-Suk Bae, Jung Kim, and Joo-Youn Jung deserve mention.  I enjoyed Jung and Jin-Suk hyung’s company at the East Asia Institute.  I thank Jung hyung and Joo-Youn unni for being generous enough to listen to me ramble on about my premature project ideas early on.  I am particularly grateful to Joo-Youn unni who provided a patient ear and warm counsel from Edmonton when I experienced both academic and life disappointments.  I also want to express my gratitude to my friends, Su-Hyun Lee and Won Hyung Choi.  Su-Hyun was generous enough to accommodate me when I accidently became homeless in Ann Arbor during the ICPSR, and I thank her for the fun time there and her constant support.  I thank Won Hyung for keeping me posted on developments in Korean politics and helping me arrange an interview with his colleagues at Hangyoreh newspaper, which helped me sharpen my ideas in Chapter 4.   Finally, and most importantly, my heartfelt thanks go to my family in Korea including Hayang Lee, my best friend of 20 years.  Despite some 5000 miles distance, she rode the emotional rollercoaster with me, being my backbone throughout the long journey. I also thank my artistic siblings, Hyunyoung and Jaehyuck, who often seem puzzled about why one would want to study politics, for their constant encouragement and humour.  Lastly, but by no means at least, I am deeply grateful to my parents, Jae-Sub Lee and Wi-Won Kim. I always feel so blessed to have them as my parents – the great combination of the most loving, caring father and tiger mother.  I will never forget that my father flew all the way from Seoul to Vancouver again in a couple of months after his first visit, saying that he had not been able to get his daughter, who probably looked emotional at the airport seeing her parents off, out of his mind.  I also thank my mother who always motivates me to push myself harder and to do my best in any area of my life.  Without the love, patience, and sacrifice of my parents, any accomplishment would not have been possible.   1 INTRODUCTION   On April 1st 2007, a day before the United States and Korea reached a bilateral free trade agreement, Heo Se-Wook, one of the Korean citizens who participated in anti-FTA demonstrations made an attempt to burn himself to death.  In his note giving the reason for his suicide, he protested the unfairness of the on-going FTA talks, mentioning that the Korean government had conceded to the US in the four main sensitive issues as prerequisites for beginning FTA negotiations – (1) US beef imports, (2) medical supplies, (3) emission standards for cars, and (4) a screen quarter system1.  Interestingly however he was not occupied with any of these four businesses: he was neither a farmer nor an employee of a pharmaceutical company, which were considered two main free-trade losers; he turned out to be a taxi driver affiliated with the taxi drivers’ labour union, which will either benefit from or at worse be neutrally affected by the bilateral free trade with the US.   Trade liberalization has generated an enormous amount of domestic political conflicts between free trade advocates and protectionists over the last two centuries.  To understand patterns of public preferences over trade liberalization policy, political economists have focused on disparate effects of trade on sets of individuals within the economy.  They often have employed the logic of economic trade theories to identify winners and losers of the policy, thereby accounting for the cross-national and individual- level differences in trade preferences.  This line of research is grounded in the assumption  1 There are four preliminaries publicly disclosed as demanded by the US as prerequisites for the opening of FTA negotiations.  These include agreeing to the “four preliminary steps” required by the US.  These steps are: (1) suspending regulations on pharmaceutical product prices (October 2005); (2) easing government regulations of gas emissions from imported US cars (November 2005); (3) resuming importation of US beef (January 2006); and (4) reducing the compulsory quota, which requires South Korean cinemas to screen South Korean films from 146 days per year to 73 days.   2 that support for or opposition to free trade develops along the lines of economic interests (Rogowski 1989; Hiscox 2001, 1-46).  The assumption however poses an empirical puzzle: if conflicts develop along the lines of narrowly defined economic self-interests as argued, how is it explained that some perceived winners from free trade, as well as some of those not directly affected by free trade – e.g., citizens as consumers, including the taxi driver above – often pose themselves against free trade?   Public opinion studies suggest that it may not be exceptional that citizens that will either benefit from free trade or be neutrally affected might take a strong protectionist position, or vice versa.  While it is widely assumed that individuals basically adopt economic policy preferences that further their private interests, numerous studies on public opinion have found that immediate and tangible self-interest often plays little or no role in determining policy preferences (Sears et al. 1980, 670-684; Sears and Funk 1990, 247-271; Wolpert and Gimpel 1998, 241-262).  Indeed, self-interest fails to influence mass preferences in such policy issues as bussing, health insurance, unemployment program, the Vietnam War, and affirmative action (Sears et al. 1980, 670-684; Lau and Sears 1981, 279-302; Kinder and Sanders 1987; Kluegel and Smith 1982, 518-532; Kinder 1986, 151-171).  Even personal economic experiences appear to exhibit weak connections to electoral behaviour; that is, so- called “pocketbook” voting where voters reward parties or candidates that advance their own economic interests and punish those that threaten them is never very strong, and is often trivial (Kinder and Kiewiet 1981, 129-161; Lewis-Beck 1990).   While self-interest tends to be surprisingly unimportant when it comes to predicting public opinion, people’s perceptions of collective conditions – e.g., national economic   3 conditions – reliably influence their political attitudes (Kinder and Kiewiet 1981, 129-161). The empirical evidence has been much more supportive of the so-called “sociotropic hypothesis” – individuals pay attention not so much to their own problems and achievements as they form their attitudes about political issues, but rather to the problems and achievements of a larger “group,” including their country as a whole (Kinder and Kiewiet 1981, 129-161).  Contrary to the assumption of most economic models that people pursue only their own material self-interest, people do care about “social” goals; and public opinion is often “group-centric” (Sears and Funk 1990, 247-271; Nelson and Kinder 1996, 1055-1078).  What “groups” do here is that they change the individual utility function, which is often considered equal to material self-interest in economic theories, by widening the scope of self-interest to include that of groups they identify themselves with or by incorporating sympathies or resentments that they feel toward the social groups they see as the principle beneficiaries or victims of the policy (Nelson and Kinder 1996, 1055-1078; Conover 1988, 51; Sniderman, Brody, and Tetlock 1991).   In the literature on the political economy of free trade, the psychological perspectives have been largely disregarded; political economists have been hesitant to sacrifice the rationality assumption in a narrow economic sense2.  For them, individuals are assumed to be  2 Recently, some albeit limited attempts have been made to investigate the role played by non-material considerations in shaping trade attitudes.  Schonhardt-Bailey (2006), for example, suggests the role of ideas as a possible explanation of disjuncture between interests and trade opinions.  More specifically, she tackles the puzzle of Britain’s repeal of its protectionist Corn Laws in 1846 – the question of why political representatives enacted a policy that appeared to harm their own economic interests and also those of the mandate on which many of them were elected – and demonstrates that ideas such as “anti-aristocracy” and the inherent “morality of free trade” appealed to general public to support the repealing the Corn Laws (Schonhardt-Bailey 2006). Hainmueller and Hiscox (2006) also point out the importance of an ideational mechanism that hinges on the role played by economic ideas and knowledge in the formation of trade preferences (Hainmueller and Hiscox 2006, 469-498).  More recently, Mansfield and Mutz (2009) find strong evidence that trade preferences are shaped less by material self-interest than by perceptions of how the country’s economy as a whole is affected by trade (Mansfield and Mutz 2009a, 425-457).  They also suggest that the empirical finding that support for trade restrictions is highest among individuals with the lowest levels of education may primarily reflect the fact that   4 rational information processors who carefully review the available information about trade policy, and then make political judgments based on acquired, processed knowledge.  While perfect economic rationality operates only under conditions of perfect information, the costs and benefits of trade policies are however not easy for laypersons to identify in advance especially when the increasing economic interdependence between states complicates personal cost-benefit calculus of the economic ties.  When the personal interest that individuals see at stake in the policy is neither clear nor high, non-material considerations likely come into play.  To compensate for shortfalls in information, individuals may need to rely on information shortcuts, such as expert advice, ideologies, party cues and campaigns. It is also important to note that individuals are not free of “impersonal influence,” i.e., individuals gather information about the world through abstract representation of anonymous mass collectives (Mutz 1998).  Under the condition that mass media facilitate influence of anonymous others, individuals may regard prospective economic gains and losses of the nation as a whole, or domestic social groups they belong to or identify with, as more important appraisals than predicted private gains and losses.   By applying theoretical findings from public opinion studies to the topic, this dissertation aims to shed new light on the question of why some individuals and some countries are more protectionist than others.  More specifically, by addressing the gap between the predictions of economic theories and the findings of public opinion studies, I examine non-economic determinants of individual attitudes toward trade policies, and institutional factors as an intervening variable that could mitigate or exacerbate protectionist  the less educated are more prone to anxieties about involvement with out-groups in their own country and beyond.    5 sentiment.  While this dissertation consists of five separate papers each of which tackles a different theoretical or empirical puzzle, it is under one overarching purpose: to demonstrate the significance of considerations of “sociotropic fairness”3 in shaping protectionist sentiment among the mass public, which will be discussed in detail in the following sections.  This introductory chapter discusses some of the major themes emerging from this dissertation.  Before introducing the themes of the ensuing chapters, I will briefly discuss why an understanding of public attitudes about trade is crucial to the politics of trade policy.  Why public opinion?  Only scant attention has been paid to the topic of mass attitudes toward trade.  The literature on the formation of trade policy relies heavily on Olson’s (1982) theory of collective action (Olson 1971; Gilligan 1997).  Individuals are assumed to engage in a cost-benefit calculation about the expected benefits of their political action; and the calculation includes both the benefits of achieving their ideal trade policy and the costs of collective action for transferring their preferences into policy (Alt et al. 1996, 689).  As a consequence, trade policy is often regarded as a function of the size of the lobbying resources spent by contenting organized interests to influence policymakers (Mayer and Li 1994, 59-77; Grossman and Helpman 1996, 265).  Given the standard trade policy analyses based heavily on influence and lobbying, it is not surprising that the mass public’s potential to shape trade policy has been understated, and consequently that public opinion has rarely been of scholarly concern.   3 I borrowed the term from Mutz and Mondak (1997); and they define it as “people’s concern with whether economic gains and losses have been distributed equitably among the nation’s many groups” (Mutz and Mondak 1997, 284-308).   6  It is not so clear that collective action problems indeed preclude public influence over trade policy, however.  Since Page and Shapiro’s (1983) classical article, “Effects of Opinion on Policy,” a number of scholars have delved into the question of how much impact public opinion has on public policy (Page and Shapiro 1983, 175-190; Stimson, MacKuen, and Erikson 1995, 543-565; Erikson, Wright, and McIver 1993; Monroe 1998, 6; Bartels 1991, 457-474).  Many of the scholars made a strong case that the public’s influence over policy is fairly substantial.  Erikson, MacKuen, and Stimson (2002) find that elected politicians consistently respond to shifts in public opinion (Stimson, MacKuen, and Erikson 1995, 543-565); Erikson, Wright and McIver (1993) also argue that state politics are responsive to state public opinion by showing that state partisanship is a strong predictor of state election outcomes, and state policies accordingly (Erikson, Wright, and McIver 1993); other studies show that public opinion influences specific policies such as environmental regulations (Hays, Esler, and Hays 1996, 41), defence spending (Bartels 1991, 457-474), and welfare spending (Fording 1997, 1-29).  Recently, after an extensive review of 30 publications in major journals, Burstein (2003) concludes that the impact of public opinion is substantial.  He finds that three quarters of the relationships between public opinion and policy among those under review are statistically significant (Burstein 2003).  Interestingly, his review shows that even when the effects of interest organizations, political parties, and political and economic elites are taken into account, the impact of public opinion remains strong.   Admittedly, demonstrating the government’s responsiveness to the public in many policy issues as above does not suffice to conclude that public opinion matters for trade policy as well.  However, given that the impact of public opinion on policy increases as   7 salience increases (Burstein 2003, 29; Kollman 1998a), it is worth consideration that trade policy has become increasingly more salient among the public over the last few decades.  For example, Johnston et al. (1992) show that the issue of free trade is of overriding importance to the 1988 Canadian national election (Johnston et al. 1992).  Likewise, the Korea-US Free Trade Agreement (KORUS FTA) was a highly publicized issue in Korea; and indeed, the decision to lift the ban on the US beef, an issue intertwined with the FTA talks, led to widespread mass protests and consequently damaged the President’s popularity and boosted the position of opposition groups, thereby enabling them to successfully block a vote on the FTA from being taken in the National Assembly (Jurenas and Manyin 2010).  Even in the United States where trade policy is expected to be not as salient as that in other countries given its unusually low levels of tariffs and trade-dependency (Kono 2008, 1224), the North American Free Trade Agreement (NAFTA) was a highly salient issue to the American public (Kollman 1998b).4  The public salience and government’s responsiveness as its consequence are not confined to the bilateral trade agreements: Kono (2008), by employing cross-national survey data, demonstrates that public support for trade liberalization leads to lower tariffs in most democracies (Kono 2008, 1224).   The significance of public opinion in the formation of trade policy necessitates exploring the determinants of individual attitudes toward trade policy.  Why are some individuals and some countries more protectionist than others?  What explains the disjuncture between self-interest and opinions?  If not or not only self-interests, what else  4 Kollman (1998) shows that interest groups that advocate popular positions on NAFTA used outside lobbying strategies to influence policymakers; and that public opinion was a crucial conditioner of the interest groups’ efforts to signal salience to policymakers.  Here, interest groups simply play a role in enhancing government’s responsiveness to the public, which is not very different from what Burstein (2003) finds.   8 determines political conflict patterns surrounding trade issues?  Also, what contextual, institutional factors, if any, explain the country-level differences in protectionist sentiment?  Distributional judgments  The recent article by Norton and Ariely (2011) presents an interesting observation: the overwhelming majority of Americans (92%) prefer the income distribution of Sweden over that of the United States (Norton and Ariely 2011, 9).  This finding is more interesting when taking into account the pervasive notion that support for welfare state runs weaker in the U.S. than in most Western democracies.  Norton and Ariely also find that this overwhelming preference for Sweden’s far more equitable distribution is robust even across all different levels of income groups.   A large body of behavioural research provides evidence that individuals like transfers that reduce inequality; and that individual policy preferences are sensitive to inequality considerations (Fehr and Schmidt 1999b, 817-868; Bolton and Ockenfels 2000, 166-193; Hausman and McPherson 2006).  Unlike most economic models assuming that people pursue only their own material self-interest, psychological evidence suggests that individual welfare depends not only on his or her own economic conditions but also the economic conditions of all other individuals in the society (Hausman and McPherson 2006; Amiel and Cowell 1999).  That is, an individual’s state of well-being is highly sensitive to social comparison and relatively insensitive to absolute income (Tversky and Griffin 1991; Esterlin 2000; Rabin 1998).  Plenty of scholarly works on ultimatum and dictator games in the experimental economics for example find that there are widely held and commonly   9 understood notions of fairness that constrain individual’s self-interested behaviour.  Given that the inequality-aversion – whether it stems from envy, insecurity, or pure altruism – is one of the most important human traits, often resulting in policy preferences deviated from self-interest based policy preferences by homo economicus, it is not hard to predict that the notion of “fairness” plays a potentially powerful role in the politics of trade policy.  Indeed, the allegedly detrimental effects of economic integration on what Kapstein (2006) calls “communitarian model of economic justice” comprise the predominant concerns with trade policies (Kapstein 2007).  By “communitarians,” Kapstein refers to those that emphasize domestic social, economic distributional implications of economic integration – i.e., issues such as the effects of greater openness on domestic income distribution and poverty and the capacity of the states to preserve domestic distributive justice (Kapstein 2007, 7).  Although Kapstein does not attempt to link the communitarian critique and public attitudes toward trade, it is not hard to predict that the domestic social implications of trade play a significant role in shaping trade opinions.  Indeed, that trade liberalization exacerbates domestic inequality and poverty has been a staple of anti- globalization rhetoric.  Among economists, however, few economic policies command as much consensus as the welfare benefits of trade liberalization.  Most economists deem it necessary to lower tariffs to enhance efficiency of resource allocation and thereby to promote economic growth of a country.  While pointing to the long-term benefits of trade liberalization – notwithstanding trade-induced adjustment costs which they argue are accrued in a short run at best – economists often view the prevailing protectionism among the public as a welfare-   10 worsening, economically costly action derived from ignorance or irrationality of the public (Caplan 2002, 433-458; Caplan 2007; Hainmueller and Hiscox 2006, 469-498).  Caplan for example argues that citizens systematically overestimate the benefits of economic protection, which leads them to vote for protection although they in fact prefer the actual effect of free trade (Caplan 2007).  Hainmueller and Hiscox likewise suggest that exposure to the economic ideas and information primarily shapes individual attitudes toward trade liberalization (Hainmueller and Hiscox 2006, 469-498).  The first two chapters of this dissertation examine how the “communitarian critique” influences individual preferences over trade policy, specifically by tackling the following questions.  Chapter 1 provides an empirical answer to the question of how much perceived domestic inequality is tolerated as a concomitant of income growth generated by trade liberalization.  As discussed above, like many economic policies, there is a trade-off between equality and efficiency in pursuing trade liberalization: trade liberalization promotes economic growth; but the trade-induced economic growth is often (or is often perceived to be) accompanied by negative distributional consequences.  Do the findings of behavioural economists that an individual’s state of well-being is affected by changes of other people’s income, and that people tend to prefer equalizing welfare gains than maximizing total welfare gains apply to the context of trade policy?  By tackling this question, the chapter proposes and tests the hypothesis that even if people are informed that trade liberalization enriches the nation as a whole, distributive consequences of the trade policy could make them feel worse off, thereby leaning them toward protectionism.    11 In a similar vein, Chapter 2 examines the sources of the female protectionism.  It has been shown that women are less likely than men to support trade liberalization; and this gap remains even after individuals’ socio-economic characteristics are controlled for.  Chapter 2 attempts to solve the puzzle of the gender gap in protectionist sentiment by drawing on the insights of the public opinion scholars, which have long suggested that public opinion is often guided by the sympathies and resentments of people feel toward the beneficiaries or victims of the policy, and the findings of the experimental economists that women are more sensitive to social cues.  More specifically, Chapter 2 proposes and tests the hypothesis that it may be gender differences in degree of sympathy for those social groups implicated in the policy that generates the gender gap in protectionist sentiment.  As does Chapter 1, this chapter also demonstrates that it is not economic illiteracy but different responses to distributive consequences of trade that account for the gender gap in protectionist sentiment.  Institutional factors  The cross-national survey data on public attitudes toward trade have revealed one interesting empirical pattern: individual skills are negatively correlated with protectionist sentiment in most developed countries, while the correlation is weak or even positive in many less developed countries (LDCs) (Scheve and Slaughter 2001, 267; Mayda and Rodrick 2005, 1393; O'Rourke et al. 2001, 157-206; Tavares 2008, 163-168).  This pattern has been widely accepted as a strong support for the economic trade theory – i.e., the Heckscher–Ohlin (Stopler-Samuelson) theorem – which postulates that economic integration has a disparate impact on countries conditional on a given country’s factor endowment.  That is, workers and the poor prefer lower tariffs in capital-scarce countries (i.e., LDCs, by definition), and   12 higher tariffs in capital-abundant countries.  The H-O logic has also served as a theoretical underpinning of the relationship between democracy and trade openness in recent studies (Tavares 2008, 163-168; Milner and Kubota 2005, 107-143; Milner and Mukherjee 2009, 163- 181).  These studies suggest that the emergence of democracy fosters trade openness in LDCs because democratization in LDCs will broaden the franchise to the wider population, most of whom are low-skilled workers that the H-O predicts are most likely to gain from trade liberalization (Mayer 1984, 970-985; Yang 1995, 956-963).   The H-O based explanations however are incomplete in the sense that the analyses are basically apolitical – that is, political aspects of the story are absent from the analyses.  It is an oversimplification to assume the socioeconomic status, such as “workers and the poor,” dictates individual preferences; and moreover, it is problematic to disregard the fact that “political agency” plays a significant role in shaping mass opinion about a proposed policy.  First, as discussed in the earlier sections, mass attitudes toward trade policy do not simply reflect the material interests that people see at stake in the issue.  The distributive effect of trade liberalization is not “socially” costless: people do care about how trade liberalization influences the domestic income distribution and poverty – e.g., who gains less and who gains more and whether or not it widens income disparity – all of which influence policy preferences.  Second, and related, public perceptions of government institutions as an agency to provide public goods and distribute the economic pie are hardly irrelevant to understanding support for economic policies.  Government is an agent to formulate and implement trade policies and provide social insurance to cushion the adverse domestic effects of the policies; and thus, the government’s desire and capacity to respond to public demands for effective policy implementation and equitable distribution of the economic pie   13 likely influence public preferences over the policies.  By tackling these points, the remaining three chapters of the dissertation provide empirical tests of the importance of government credibility in determining public support for trade liberalization policy, and discuss their implications for new democracies.  Chapter 3, “The Compromise of Embedded Liberalism and Government Credibility,” provides evidence of effects of government credibility on protectionist sentiment.  According to the new embedded liberalism thesis by John Ruggie, economic openness and welfare are mutually reinforcing.  The existing literature exploring the micro- foundation of the openness-welfare nexus suggests that individuals’ perceptions of economic insecurity lie at the core of the thesis – i.e., openness creates economic insecurities, which in turn inspire demand for welfare policy as compensation.  This argument is, however, only half the story of the nexus: demand side.  The embedded liberalism thesis involves a “grand bargain” whereby the public agrees to open markets and a government in exchange promises to moderate the volatility of open markets and provide social safety nets as compensation (Abdelal and Ruggie 2009).  For this grand bargain to be enacted and sustained, the public should be willing to compromise on trade liberalization policies in return for the government’s welfare promise.  A problem of credible commitment arises here, because trade liberalization and a government’s compensation often involve inter-temporal, non- simultaneous exchanges between the government and the public.  Government attempts to invest in social welfare in the hope that its investments help build public support for openness; but compromise of any significance will not be made unless the government commitment to welfare expansion is viewed by its public as credible.  Likewise, there is not much incentive for the government to increase welfare spending, unless welfare spending is   14 seen as a viable solution to the political dilemma faced by democratically elected politicians who are committed to trade liberalization.  Social factors such as the public’s confidence in strong, effective governmental institutions to protect them against vagaries of free market economy, however, take time to develop as it requires government to engage in repeated transactions with the public (Kapstein and Converse 2008).  Given such temporal feature of the grand bargain, people are likely to compromise only when the government’s promises to cushion the adverse domestic effects of open markets are viewed as credible.  If viewed as not credible, the grand bargain would not be enacted, let alone sustained.   The findings from Chapter 3 suggest that it is highly likely that material compensation itself that is directed toward losers from trade has no discernable economic insecurity-reducing functions in countries where there is little trust in government.  It is government credibility that is essential to offsetting trade-induced economic insecurity, but such credibility takes time to build.  If the reinforcing relationship between trade and welfare state can only be long-term and historically contingent, LDCs that have never enjoyed the privilege of expansive social protection and are often characterized by lack of effective government institutions, insecurity-reducing functions of government promises are not likely to work at the same level as they do in countries with an effective government and a long welfare state tradition.  By taking up this point, Chapter 4 further investigates the effects of political trust on mass attitudes toward trade liberalization policies in new democracies in more detail, using the case of Korea.   As an export-oriented economy where trade has played a major role in boosting the country’s rapid economic growth over the last few decades, Korea is one of the countries   15 displaying a strong public consensus on the positive impact of trade on national economies (Chicago Council on Global Affairs, 2002, 2004, and 2007).  However, when it comes to the government’s specific trade policy initiatives, such as the Korea-US free trade agreement or the EU-Korea free trade agreement, the public’s support is considerably lower.  By tackling the disjuncture between a willingness to accept increased international trade in general and a hesitation to support specific trade policy initiatives, this chapter hypothesizes and demonstrates that a lack of “political trust” – defined here as public evaluations toward the responsiveness of the political process and effective governance – is an important source of the divergence.  I argue that if the public feels that the policymaking process is not responsive and transparent enough to make their voices heard, and that government institutions in charge are incapable, ineffective, and inefficient to carry out their responsibilities, just as the Korean cases illustrated in this chapter, then the public should distrust and reject its policies, especially those which involve a great deal of uncertainty and risk, such as trade policies. While the research focuses on the case of Korea, the results suggest that the causal significance of political trust in trade policy preferences is likely to be generalizable to most new democracies that are characterized by weak political institutions and ineffective regime performance.   The findings of Chapter 3 and 4 however are at odds with the H-O based model exploring between democratization and trade liberalization.  The H-O model combined with Duncan Black’s median voter theorem support the argument that democratization positively influences trade liberalization.  Given that democratization by definition implies transferring power from non-elected elites to the wider population, and most of the electorate – and notably the median voter – are relatively capital poor in LDCs expected to benefit from   16 trade, democratization is likely lead to more economic openness (Milner and Kubota 2005, 107-143; Milner and Mukherjee 2009, 163-181; Milner and Mukherjee 2009, 163-181; Mayer 1984, 970-985; Yang 1995, 956-963; Kono 2008, 942-955).  Basically, the theoretical argument about democracy and trade policy is built on the two assumptions: (1) the electorate prefer more trade openness in LDCs and (2) democracies are more responsive than autocracies to electoral pressures.  Given the findings of Chapter 3 and 4, however, the first assumption does not seem to hold.  The electorate in economically less-developed and politically nascent democratic countries do not seem to be necessarily in favour of trade liberalization – rather, it seems more plausible that they are more likely to be protectionist than those in developed and advanced democratic countries.   The last chapter of this dissertation thus provides an empirical test of the H-O assumption that the electorates (the median voter) in LDCs more likely prefer lower levels of protection than those in developed countries.  I find that at the country level, protectionist sentiment is significantly high among people living in capital-scarce countries, countries with high level of economic insecurity, and countries with a strong public demand for income redistribution.  These findings imply that trade attitudes are not simply a function of who is personally benefited or hurt by trade in the labour market, conditional on a factor endowment proportion of a given country.  Rather, it suggests that economic insecurity and perceived social injustice (income inequality) can be an important motivation behind protectionist pressures, more so than perceived material benefits of trade.  Like the findings of the earlier chapters, the findings of this chapter confirm that unlike the H-O prediction, protectionist forces more likely prevail in new democracies that are often characterized by   17 weak political institutions and ineffective regime performance, than in advanced democracies.   18 1 DISTRIBUTIONAL JUDGMENT IN INDIVIDUAL PREFERENCES OVER TRADE LIBERALIZATION   In justice as fairness, men agree to share one another’s fate – Rawls (1971)  1.1 Introduction  Mainstream economists evaluate economic policies or their outcomes by asking whether they make the people better off.  They often assume that one’s well-being depends on her material well-being – i.e., that one can improve well-being by increasing her own income; and that public policies aimed at increasing the income of the nation as a whole lead to greater public well-being (Easterlin 1995, 35-47).  When the economists’ evaluation of trade liberalization policies is contrasted with public preferences over the policies, this assumption poses an interesting puzzle.  Few economic policies command as much consensus among economists as the welfare benefits of trade liberalization; yet public opinion polls have consistently revealed that the public in most industrialized countries has been largely sceptical of trade liberalization policies (Caplan 2007; Irwin 2005a; Fuller and Geide- Stevenson 2003, 369-387).  What explains the discrepancy in views on trade liberalization between economists and the public?   For mainstream economists, public preferences over economic policy are conventionally understood as an expression of the material interest that individuals see at stake in the policy.  Indeed, economists presume that the policy opinions are expected to be divided along the lines of winners and losers of free trade; and thus have focused their attention on identifying how trade liberalization is expected to affect the individual economic   19 welfare of citizens.  Moreover, while pointing to the long-term benefits of trade liberalization – notwithstanding trade-induced adjustment costs which they argue are accrued in a short run at most – many tend to attribute the prevailing protectionism among the public to a mere reflection of ignorance or irrationality of the public (Caplan 2002, 433-458; Caplan 2007; Hainmueller and Hiscox 2006, 469-498).  Caplan for example argues that citizens systematically overestimate the benefits of economic protection, which leads them to support protection although they in fact prefer the actual effect of free trade (Caplan 2007). Hainmueller and Hiscox likewise suggest that exposure to the economic ideas and information primarily shapes individual attitudes toward trade liberalization (Hainmueller and Hiscox 2006, 469-498).   In this paper (also Chapter 1 of the dissertation) however I argue that citizens vote for protectionism not primarily out of their low levels of economic literacy; but because they do not prefer the alleged effects of trade: inequality and poverty.  Indeed, trade produces both positive and negative externalities: trade enhances efficiency and promotes economic growth; but at the same time it creates distributive consequences.  As Hausman and McPherson (2006) clearly note, it is not generally possible to separate questions concerning efficiency from distributional questions – i.e., between the size of the pie and the way it is sliced (Hausman and McPherson 2006).  They argue, by citing Samuelson (1950), even a preference for policies that generate Pareto improvement – i.e., a change in income distribution that makes at least one individual better off without making any one else worse off – is not in general free of distributional commitments (Samuelson 1950, 1-29 and Hausman and McPherson 2006, 147).  Given that these positive and negative externalities are intertwined with one another thereby affecting trade preferences, the prevailing protectionism among the   20 public may not stem from lack of understanding of the positive externality, but from strong countervailing effects of negative externalities against the positive.  While the mainstream economists tend to focus on the distribution of private costs within society as a basis of political opposition to trade liberalization, the modern behavioural economists find that individual’s utility function is dependent not only on his or her own economic conditions but also the economic conditions of all other individuals in the society (Hausman and McPherson 2006; Amiel and Cowell 1999).  That is, trade liberalization may not be socially costless even when it increases the total income of the society without making anyone worse off.  Given that one’s well-being is affected by changes of other people’s income and that people tend to prefer equalizing welfare gains than maximizing total welfare gains (Easterlin 1995, 35-47; Merton 1957; Easterlin 2000, 18-19), the distributive consequences of trade likely affect individuals’ preferences over trade liberalization – not only that of those directly affected, but also that of those relatively unaffected by trade.  Against this background, this chapter aims to demonstrate how much perceived inequality and poverty are tolerated, as a concomitant of income growth generated by trade liberalization.   More specifically, this paper tackles the following issues.  First, I examine if the findings of behavioural economists that an individual’s utility is highly sensitive to social comparison and relatively insensitive to absolute income (Tversky and Griffin 1991; Easterlin 2000; Rabin 1998) apply to the context of trade policies.  Basically, I challenge the Pareto principle – which suggests that if at least one person is made better off and no one is worse off, then the social welfare must increase – by demonstrating that even if trade   21 liberalization enriches the nation as a whole making no one worse off, distributive consequences of the policy could lean them toward protectionism.  Second, I examine how much (perceived) trade-induced poverty would affect individuals’ trade policy evaluations.  I demonstrate that even if trade liberalization has positive effects on a country’s economic growth, individuals would not prefer the policy if it means that the poor (the vulnerable) needs to be sacrificed (at least for the short term) for the general good of national income growth.  Finally, I demonstrate that the Rawlsian preferences – inequality averse and other- regarding preferences – remain strong even after controlling for levels of economic literacy or feelings of job prospects.   This paper proceeds as follows.  In the next section, I discuss the theoretical questions I explore and propose specific hypotheses to test.  The focus is on how trade- induced positive and negative externalities offset each other and thereby affect individual trade preferences.  I next describe the design of the studies I conducted and analyzed. Finally, I describe the experimental results and outline their implications.  1.2 Externalities of trade and trade preferences  Trade produces externalities that influence the country either positively or negatively. Externalities or spillover effects from trade policies mean benefits or costs that accrue to a third party who is not directly involved with the relevant transactions – i.e., benefits or costs that influences the standard living of society as a whole while they are not transmitted through prices to the parties to the relevant transactions.  Economists have long believed that trade liberalization enhances economic efficiency and promotes economic growth in the country.  Trade may also produce negative externalities, such as inequality, insecurity, and   22 poverty.  This study focuses on two externalities of trade, i.e., the effects of trade on inequality and poverty, and examines to what extent individual perceptions of the positive externality, i.e., growth, are offset by the perceptions of the two negative externalities in determining trade preferences.  This section introduces definitions of the concepts and hypotheses put to the test. 1.2.1 Positive externality: Growth  Schonhardt-Bailey (2006), while examining the role of ideas in Britain’s repeal of its protectionist Corn Laws, demonstrates that ideas of positive externalities, i.e., ideas couched in terms of the broader social welfare, can be used to garner support for trade liberalization (Schonhardt-Bailey 2006).  Among potential positive externalities that trade liberalization may generate, that trade enhances economic efficiency and growth is probably the most appealing to a wide audience.  Indeed, the idea that the open economies lead to more efficient allocation of resources and thus fare better in aggregate than do closed ones is very old, going back to Adam Smith and David Ricardo.  More recently, economists have demonstrated trade liberalization can expedite the rate of growth through its incentive effects on investment, spillovers of technology, and innovation (Romer 1989; Grossman and Helpman 1992; Hejazi and Safarian 1999, 491-511).  Empirical studies largely support that trade is correlated with, and often a source of, growth (Dollar 1992, 523-544; Sachs and Warner 1995; Sachs and Warner 1995).  Scholars exploring public preferences for trade have recently suggested that the positive impact of education on trade attitudes may be explained by levels of exposure to the economic idea – i.e., the idea that trade induces the overall efficiency gains and growth for the national economy (Hainmueller and Hiscox 2006, 469-   23 498; Caplan 2006, 367-381).  I first test the impact of the positive externality on trade preferences.  The specific hypothesis with relation to growth is as below:  H1: The idea that trade liberalization brings about economic growth will increase individuals’ support for trade.   The positive externality is hardly the only concern for individuals in forming their policy attitudes, however.  Admittedly, mainstream economists and policymakers tend to equate social welfare with national efficiency gains or growth; but efficiency gains by trade liberalization are often accompanied by distributive consequences within countries, which likely comprise negative externalities for individuals.  If individuals care not only about how trade policy influences absolute income growth but also how it affects their income growth relative to others, or social inequality in general, with a preference for policies that promote income equality, they may not support the policy even when they are aware of the positive externality of trade.  In the following section, I will discuss potential effects of negative externalities of trade on individual trade preferences.  1.2.2 Negative externality: Inequality  That trade liberalization exacerbates domestic inequality has been a staple of anti- globalization rhetoric.  Admittedly, scholars do not seem to have reached a consensus on the question of whether it is indeed trade liberalization that contributes to widening income inequality being observed in many countries.  Scholars such as Lundberg and Squire find that trade liberalization has led to a pattern of growth that disproportionately benefits the rich thereby worsening income inequality (Lundberg and Squire 1999).  Other scholars however   24 cast doubt on this assessment, showing that trade benefits the poor to the same extent that it benefits the whole economy (Dollar and Kraay 2000; Freeman 1995, 16-21). Notwithstanding the controversy, the simultaneity of the two developments seems sufficient to make economic openness the most prominent culprit in the eyes of the general public. Given that it is perception of the externalities that matters in shaping trade preferences, it is probably more important to ask how people perceive the link between trade and inequality, rather than whether trade indeed exacerbates inequality or not.   Inequality is a complex concept.  It is a concept that should be distinguished from insecurity; and feelings of total altruism or other-regarding behaviour.  First, I argue that inequality and insecurity are two different things.  In fact, economic insecurity – job insecurity and labour market insecurity – has been considered the most frequent and powerful rationale for protectionist sentiment (Irwin 2005a; Hays, Ehrlich, and Peinhardt 2005, 473), while inequality in the existing literature has either been treated interchangeably with insecurity or attracted scant attention.  Indeed, since Atkinson’s pioneering work, inequality and insecurity have often been treated as “intellectual cousins” (Amiel and Cowell 1999; Atkinson 1970, 244-263).  Harsanyi, for example, used individuals’ insecurity as basis for individuals’ inequality-averse attitudes (Harsanyi 1953, 434).  However, I argue that the perceptions of insecurity and inequality in the context of trade liberalization, though closely related, differ at least in the following two senses.  First, concerns about increasing inequality inherently include perceptions of how trade liberalization affects people at the aggregate – the so-called “sociotropic” perceptions – whereas concerns about insecurity are in principle egocentric in the sense that changes in others’ income do not need to be a part of the concern.  Anderson and Pontusson (1997), for example, conceived of economic insecurity as a function of both   25 (1) the individual’s estimate of the probability that she will lose her job and (2) the individual’s perception of the consequences of losing their job (Anderson and Pontusson 2007, 211-235).  In contrast, perceptions of inequality are closely associated with interpersonal comparisons, and could be viewed as shaping attitudes toward trade liberalization, irrespective of increase or decrease in personal gains.  Second, and related, non-negative income change – i.e., income increases that accrue to everyone in the society – could invoke inequality perceptions, yet it does not evoke feelings of insecurity.  For instance, more income to one person and no income loss to anyone else could increase the perception of inequality (if the income increase happens to go to the rich) but not that of insecurity.   Second, I use the concept of inequality to refer to “reference dependence” of an individual’s utility function, which is differentiated from an individual’s other-regarding preferences which I will discuss in the next section.  By reference dependence, I mean that the tendency that people evaluate a policy not only by calculating the policy-induced increase in absolute income but also by making relative utility comparisons with a reference group, e.g.., “the rich”.  Suppose that one segment of the society were made better off, but on this occasion it happened to be the richest in the community.  If that is the case, the relative position of the lower class worsens as average income rises.  Although this sort of non- negative income change increases total income of the society with no one made worse off, some people might not feel that this represents an increase in social welfare.  Indeed, scholars have empirically demonstrated that people tend to prefer equalizing welfare gains rather than maximizing total welfare gains (Amiel and Cowell 1999; Rabin 1998, 11-46); and that the Rawl’s difference principle (Rawls 1999)– i.e., in order for any change to be accepted as an improvement, the change should be the one that increases the utility of the worst-off   26 individuals in society – best describes individual preferences when there is trade-off between efficiency and equality in pursuing economic policies (Mitchell et al. 1993, 629).  The Rawlsian idea has also been tested in the context of trade liberalization.  Herrmann et al. (2001) find that a large percentage of the general public are intuitive Rawlsians (Herrmann, Tetlock, and Diascro 2001, 191-218): their experiments reveal that nearly 46 percent of American public favour restricting trade when the benefits go to the rich.   This paper focuses on the question of to what extent individuals’ trade policy evaluations are affected by the perceived distributive consequences of the policy, as a concomitant of income growth generated by trade liberalization.  By tackling the psychological implications of the trade-off between efficiency of income generation (economic growth) and equality of income distribution, this paper attempts to demonstrate that (1) the assumption that people’s policy preferences are shaped primarily by income growth is not substantiated; (2) protectionist backlash therefore reflects not so much that people are irrational as Caplan (2007) argues or lacking sufficient economic knowledge, but that they have inequality-averse social preferences, as an effective countervailing force against the effect of aggregate benefits from free trade; and (3) perceptions of job insecurity, measured by perceptions of job prospects here, shape trade policy preferences independently from inequality-averse preferences5.  The specific hypotheses to be tested in relation to inequality and insecurity are as follows:   5 In fact, scholars have suggested that a lack of job security could lead students to make status quo choices with their career: for example, Chang (2011) argues that the surging popularity of medicine as a subject among university applicants in Korea can be accounted for by the dramatic fall in job security after the 1997 financial crisis (Chang 2011).   27 H2: Even if trade liberalization enriches the nation as a whole, many individuals will not prefer the policy if it widens income disparities between classes.  H3: Inequality-aversion has little to do with levels of economic knowledge: even after controlling for economic knowledge or economic education experience, the effect of inequality-aversion on trade preferences will remain strong.   H4: Concerns about their job prospects will be negatively associated with support for free trade, independently from inequality-averse preferences.  1.2.3 Negative externality: Poverty  Poverty and inequality are two different things.  They touch on issues in common, but the presence of inequality does not necessarily mean the presence of poverty.  In discussion of inequality above, for example, I only considered the case that trade liberalization makes all individuals in the society better off, while income goes disproportionally to the rich.  In this case, one can say trade liberalization makes inequality worse but not poverty worse.  This distinction is important because inequality and poverty discussed here draw upon different ethical concerns.  Concerns for the poor – the least privileged citizens – are different from concerns for individuals’ own relative gains, which is summarized as “envy” or “insecurity” discussed above, in the sense that they are derived from “other-regarding” preferences.  John Rawls’ justice in fairness argument that in the original position under a veil of ignorance, “men [will] agree to share one another’s fate” (Rawls 1999) offers an ethical viewpoint to explain why individuals often depart from pure self-interest as narrowly defined to pursue other-regarding goals.   One of the biggest concerns of critics of trade liberalization is its impact on the poor. Scholars however do not seem to have reached a consensus on the issue of whether trade   28 liberalization is good or bad for the poor.  Most economists posit that trade promotes economic growth, which consequently helps alleviating poverty (Bannister and Thugge 2001; Winters, McCulloch, and McKay 2004, pp. 72-115). They argue that trade liberalization contributes significantly to poverty alleviation by improving resource allocation in favour of efficient activities where a given country enjoys comparative advantage, and making cheaper products more available to local consumers.  Others however point out that the adjustment costs of trade liberalization often harm poorer actors in the economy, which could leave them behind in poverty even in the long term.  Notwithstanding the controversy over the time horizon, it is hard to refute that those adversely affected by trade liberalization can be disproportionately poor at least in the short term, because the poor have fewer assets to absorb adjustment costs during economic hard times than other segments of society.  Indeed, criticisms of trade liberalization’s impact on the poor abound despite the general agreement of the economic literature that open economy promotes growth and growth is good for the poor (Harrison 2007).  Along with inequality-aversion discussed above, this paper aims to see how much (perceived) trade-induced poverty exerts a countervailing effect on trade preferences against the positive externality of trade (growth).  The specific hypotheses to be tested are as follows.  H5: Even if people are informed that trade liberalization has positive effects on economic growth, many individuals will oppose the policy if they believe that implementation of the policy  means that the poor (the vulnerable) needs to be sacrificed for the general good of national income growth.   H6: The effect of other-regarding behaviour on trade preferences will remain strong even after controlling for levels of economic knowledge.     29 1.3 Method: Survey experiments  This paper uses a survey experiment to directly test the hypotheses above.  The study was conducted in April 2011 at the Public Opinion Laboratory at the University of British Columbia.  A total of 91 undergraduate students from the political science subject pool participated in the study.  The experiment is a computer-based test designed to study effects of different treatments on responses.  For this experiment, the subjects were randomly assigned to four different groups.  Except for a control group receiving no introduction, each group received different introductions to a survey question about international trade. These introductions mention effects of trade on national income growth, income distribution, or poverty.  The exact wordings are shown below:  Group 1: Control (No introduction)  Group 2: Growth Suppose you believed that removing most restrictions on foreign imports would increase economic growth.  Group 3: Growth and high-income qualifier Suppose you believed that removing most restrictions on foreign imports would increase economic growth, but that it would mostly benefit people with high incomes and would provide little or no benefit for others.  Group 4: Growth and poor people qualifier Suppose you believed that removing most restrictions on foreign imports would increase economic growth, but that it would generally hurt low-income workers.    After the introductions were given, all subjects were asked the same core question about their attitudes toward government restrictions on foreign imports, which are presented below.  The specific language was chosen because “restrictions on foreign imports” is the most widely used language to measure protectionist sentiment in public opinion surveys, such as World Value Survey, Asia-Europe Survey, or International Social Survey Programme.   30 In addition, in order to avoid forcing subjects to choose between an extreme view and all other less extreme views, I used the word “most” to make the policy less extreme – that is, from “remove restrictions” to “remove most restrictions”.  (For subjects in Group 1) Do you agree or disagree with this statement: Canada should remove most restrictions on foreign imports. - Agree - Disagree  (For subjects in Group 2) Would you then agree or disagree with this statement: Canada should remove most restrictions on foreign imports. - Agree - Disagree   The specific claims in the introductions provided to the different groups were designed to measure the effect of positive externality – growth (Group 2) – and to what extent trade’s positive externality is offset by its negative externalities – inequality (Group 3) or poverty (Group 4).  These introductions allow us to test H2 and H4 above. If H1 is confirmed, we will see higher support for trade in Group 2 than the control group (Group 1). If H2 is confirmed, we will see the high-income qualifier in Group 3 slide subjects toward more protectionism, in comparison with those exposed to the growth introduction (Group 2), despite the information that trade liberalization has positive effects on economic growth, making no one worse off in absolute terms.  Likewise, if H5 is confirmed, we will see those exposed to the poor people qualifier (Group 4) are more inclined toward protectionism than people in Group 2.  I also included items to measure levels of economic literacy in order to test H2 and H5 and items to measure feelings about job prospects.  I first ask a question whether   31 respondents have taken any courses in economics (econ course).  In addition, to measure levels of economic literacy directly, I included four quiz items to measure economic literacy.  The exact wordings of the four quiz items are presented below.  The three questions except for the first question are adopted from Test of Economic Literacy, National Council on Economic Education (1987), which was re-used by Walstad (Walstad 2002, 63–96).  By adding up these four items, I created an index variable economic literacy.  Along with the economic literacy measures, I also included questions asking how optimistic respondents feel about their job prospects after graduation (egocentric job prospects) and about job prospects for Canadians in general (sociotropic job prospects).  They will be used as proxies for personal economic insecurity among undergraduate students.  According to the economic theory of comparative advantage, which countries benefit from international trade?  All countries6 Small countries Poor countries Wealthy countries   Which of the following challenges do all economic systems face?  How to:  balance imports and exports balance the government’s budget make the best use of scarce resources save money to reduce the national debt   One reason the federal government might reduce taxes is to:  slow down the rate of inflation slow down a rapid rise in interest rates decrease business spending on plant and equipment increase consumer spending and stimulate the economy   6 Correct responses are shown in bold.   32  Which of the following statements about tariffs is true?  Tariffs benefit export industries. Tariffs benefit consumers. Tariffs benefit import-competing industries. Tariffs encourage the growth of the most efficient industries.  1.4 Results: Effects of inequality and poverty on support for trade  Table 1.1 reports the simple frequencies of each type of response in each of the four treatment groups.  The results show that (1) the growth introduction has large effects on responses; and that (2) both inequality and poverty introductions have significant countervailing effects against the growth introduction.  First, compared to respondents who were given no introduction to the question about trade (Group 1), 51 percent more respondents said they would agree the statement that Canada should remove more restrictions on foreign imports if it increased economic growth (Group 2).  Second, however, when the growth introduction is presented in combination with the high-income qualifier or the poor people qualifier, the effect of the growth frame weakens substantially.  Compared to those who received the growth introduction only (Group 2), 45% fewer respondents who received both growth introduction and high-income qualifier (Group 3) said they would agree with Canada removing restrictions on foreign imports.  Likewise, although to a lesser degree, the poor people qualifier has a similar effect: 23% fewer individuals agreed to removing restrictions.  In short, the results confirm the two hypotheses above (H1 and H3): even if they are informed that trade liberalization promotes economic growth, many individuals would not support the policy if it widens income disparity among classes or harms disproportionately the poor.    33 Table 1.1 Treatment effects on responses about import restrictions   Question: Would you then (or Do you) agree or disagree with this statement: Canada should remove most restrictions on foreign imports.   Removing restrictions All respondents (N = 74) Disagree 43 (58%) No introduction (Group 1: N = 15) Disagree 12 (80%) Growth (Group2: N = 17) Disagree 5 (29%) Growth and high-income qualifier (Group 3: N = 19) Disagree 14 (74%) Growth and poor people qualifier (Group 4: N = 23) Disagree 12 (52%)     Then, do the countervailing effects of high-income and poor people qualifier against growth primarily reflect lack of exposure to economic ideas, as often suggested (Caplan 2002, 433-458; Hainmueller and Hiscox 2006, 469-498; Caplan 2006, 367-381; Hiscox 2006, 755- 780)?  Table 1.2 reports the response frequencies in each treatment group subdivided by “education experience” – i.e., a variable that divides respondents into two groups: those who have taken a course in economics and those who have not.  Basically, no significant difference is found between the two groups.  Admittedly, one might argue that the education effect is not significant simply because the treatments here in essence require respondents to make choices under hypothetical conditions.  However, even when only those in the control group with no hypothetical treatments are taken into account, no education effect is found: the difference in support for trade between those who have taken courses in economics and those who have not is of no statistical significance.  In fact, the only significant difference is found among those who received the growth introduction – it appears that those who have taken a course in economics are more susceptible to the growth introduction than those who have not.  When the growth introduction is given in combination with high-income or poor   34 people qualifier, however, the education experience makes little difference: respondents who received any of the qualifiers are in general hesitant to agree to removing import restrictions to the same degree regardless of whether they have taken a course in economics or not.  This suggests that there is no evidence for value difference, i.e., difference in inequality- and poverty-aversion between those who have taken courses in economics and those who have not.  Table 1.2 Economic education and treatment effects   Question: Would you then (or Do you) agree or disagree with this statement: Canada should remove most restrictions on foreign imports. Remove restrictions Econ course taken Econ course not taken Differences All respondents (N = 74) Disagree 25 (55%) 18 (64%) -9% Control (Group 1: N = 15) Disagree 7 (78%) 5 (83%) -5% Growth (Group 2: N = 17) Disagree 2 (22%) 3 (43%) -20% Growth and high-income qualifier (Group 3: N = 19) Disagree 8 (73%) 6 (75%) -2% Growth and poor people qualifier (Group 4: N = 23) Disagree 8 (50%) 4 (57%) -7%      However, there are stark differences between those who are optimistic about their job prospects and those who are not (H4).  Table 1.3 shows that across the board respondents who are optimistic about their job prospects are substantially more likely to agree to removing restrictions on foreign imports than those who are not optimistic.  The difference is starkest among respondents who received the growth introduction only (Group 2): compared to those who said they were not optimistic about job prospects, among those said optimistic 72% more individuals agreed to Canada removing import restrictions in this group.  Likewise, albeit to a lesser degree, both high-income and poor people qualifiers exert   35 more negative influence on support for trade among those not optimistic about job prospects.  These results are basically in line with the existing empirical finding showing that trade preferences are closely associated with economic (or job) insecurity.  But it is noteworthy that it is respondents’ general lack of confidence in job “prospects” – not their current perceived job insecurity but job insecurity about their future – that shaped trade preferences.  The results have two important implications: (1) the facts that respondents who are not optimistic about their job prospects are not as much attracted to the growth introduction as those who are optimistic, and that they in general do not prefer removing restrictions suggest that a lack of confidence in job prospects may lead people to make status quo policy-decisions; and (2) the fact that the respondents are “students” from the political subject pool that are likely to be constituting a socio-economically homogenous group, suggests that the effect of job prospects is largely psychological not based on their social- economic status.  The question is, then, how do perceptions of job prospects affect the inequality- and poverty-aversion preferences?    36  Table 1.3 Job prospect and treatment effects   Question: Would you then (or Do you) agree or disagree with this statement: Canada should remove most restrictions on foreign imports. Remove restrictions Job optimistic Job not optimistic Differences All respondents (N = 71) Disagree 15 (37%) 30 (83%) -46% Control (Group 1: N = 14) Disagree 4 (67%) 7 (87%) -20% Growth (Group 2: N = 17) Disagree 1 (8%) 4 (80%) -72% Growth and high-income qualifier (Group 3: N = 19) Disagree 5 (56%) 9 (90%) -34% Growth and poor people qualifier (Group 4: N = 21) Disagree 5 (36%) 5 (71%) -25%    To gauge the impact of the treatments and its relation with economic knowledge or job insecurity in a more precise manner, I estimate individual trade preferences using binary logistic regression.  Table 1.4 reports the results from logistic estimations that include the core explanatory variables above.  The numbers in the table are marginal effects with standard errors in parenthesis.  Model 1 estimates the general impact of each different introduction to the trade question (1 = disagree to removing import restrictions (anti-trade) and 0 = agree (pro-trade)), using the “no introduction” treatment as a baseline category here. Consistent with the findings in Table 1.1, the estimated general effect of the growth introduction is to increase support for trade (removing import restrictions) by 55.1% (s.e. 20.4%), that is, having exposed to the growth introduction versus no introduction at all increases support for trade by 55.1%.  Model 2 is the same model as Model 1 but with the growth treatment as a baseline category.  Model 2 shows that having exposed to the high- income qualifier along or the poor people qualifier with the growth introduction decreases   37 support for trade by 46.4% (s.e. 18.1%) and 23.4% (s.e. 16.5%), respectively, in comparison with having exposed only to the growth introduction.   Models 3 and 4 report the results from logistic estimations that include a set of different measures of economic education.  Model 3 estimates the effect of economic education, using the variable econ education (1 = having taken a course in economics and 0 = not having taken a course), which confirms the findings in Table 1.2: having taken a course in economics has no effect on trade preferences.  Model 4 repeats the same analysis replacing the binary variable for economic education (econ education) with the index variable economic literacy on a scale from 0 to 1 where 0 means no correct answers to the economic literacy questions presented in Appendix A and 1 means all correct answers to those questions.  Model 4 shows that economic literacy has a significant positive impact on the probability of agreeing to remove trade restrictions: 10 percent increase in economic literacy corresponds to 6 percent more support for trade.  Interestingly, Model 4 also shows the impacts of the experimental treatments on trade preferences become larger when economic literacy is taken into account: the estimated effect of the growth introduction increases from 53.5% to 64.6% (s.e. 22.4%); and the estimated effects of high-income qualifier and poor people qualifier increase by 6.6% and 2.4%, respectively.  The fact that taking economic literacy influence into account does not show the three treatments to be spurious is important because it implies that the ideas that trade produces growth, inequality, or poverty indeed matter for shaping trade preferences and that the effects do not stem from lack of economic literacy.  Individual interests in learning economics however have no influence on support for trade (Model 5).    38  Model 6 and 8 show that both egocentric and sociotropic job prospects have statistically significant and substantively strong impacts on support for trade, which basically confirms findings in Table 1.3.  Compared to those who said they felt not optimistic about their personal job prospects after graduation, among those who said optimistic, 51% more people agreed to removing restrictions on foreign import; and these differences between those who are optimistic and those who are not are statistically significant.  Sociotropic job prospects, albeit to a lesser degree, have the similar effect: 28% more of those who said they felt optimistic about job prospects for Canadians said they agreed to removing import restrictions than those who said not optimistic.  Job prospects measures however do not make the effects of the three treatments any less important.  What is interesting is that feelings about personal job prospects are closely associated with economic literacy, thereby having an impact on trade preferences.  Model 7 shows that taking job prospects variable into account makes the relationship between economic literacy and trade preferences substantively less important and statistically insignificant.  This suggests that it may be not so much lack of economic knowledge as economic insecurity that leads people toward protectionism.   Table 1.4 Logistic regression: Protectionist sentiment (Individual attitudes toward restrictions on foreign import) Dependent variable = 1 if respondent disagree with the statement that Canada should remove most restrictions on foreign import  (= 0 if agrees).  Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Introductions Control   0.551** (0.204)  0.535** (0.205)  0.646** (0.224)  0.519* (0.223)  0.504* (0.228)  0.583* (0.247)  0.543* (0.213) Growth -0.551** (0.204)  Growth and high-income qualifier -0.087 (0.202)  0.464* (0.181)  0.446* (0.182)  0.512** (0.191)  0.402* (0.191)  0.450* (0.204)  0.501* (0.211) 0.479* (0.195) Growth and poor people qualifier -0.316. (0.187)  0.234 (0.165)  0.228 (0.168)  0.252 (0.171)  0.210 (0.171)  0.216 (0.190)  0.224 (0.191) 0.202 (0.180)  Economic education Econ course   -0.102 (0.131)  Economic literacy    -0.608* (0.271)   -0.489 (0.313)  Interest in economics     -0.079 (0.271)   Job prospect Egocentric     0.509*** (0.154)  0.451** (0.159)  Sociotropic         0.278* (0.140)  Log likelihood -44.676 -44.676 -44.008 -41.839 -43.825 -36.600 -35.298 -39.181 Observations  74  74  73  74  72 71 71  68 Notes: Logit estimation: marginal effects ( XXYP  /)|1( ) are shown with standard errors in parenthesis.          Significant codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1      40 The results can be summarized as follows.  First, the ideas of all three externalities trade produces – growth, inequality, and poverty – exert significant influence on support for trade. More importantly, the two negative externalities of trade liberalization – inequality and poverty – have strong countervailing effects on support for trade against the positive externality of trade (growth): that is, respondents reject trade liberalization in which the rich become richer or the poor become poorer even if it leads to economic growth.  Second, both economic literacy and economic insecurity are associated with trade preferences: the more economic literate the more support for trade; the higher economic insecurity the less support for trade.  Even when controlling for economic literacy or economic insecurity, however, the effects of the externalities on trade preferences do not become any less important.  Third, taking economic insecurity into account shows the relationship between economic literacy and trade preferences to be spurious.  This suggests that it is their job prospects rather than levels of economic literacy that shape trade preferences.  Given that respondents are undergraduate students from the political science subject pool who can be viewed as potentially constituting a socio-economically homogenous group, what appears to matter more for shaping trade preferences is a psychological factor, such as feelings about job prospects, rather than skill-levels or exposure to economic knowledge as often suggested.   Admittedly, the experiments are not without limitations.  Although the experiments reported here were designed mainly to see the impacts of “ideas” on policy preferences, the possibility that the results may also reflect framing effects – e.g., effects of different levels of intensity of language used in each treatment, which might create emotional responses – was not excluded.  Also, the internal validity of the experiments is achieved at the expense of external validity.  The data collected here represent only the opinions of the university   41 students.  Notwithstanding the limitations, it is hard to deny that equality-growth trade-offs in shaping trade preferences are strong; and that there is no compelling reason to predict for other subject populations systematic divergences from the patterns found here.   1.5 Conclusion  Recently, French president Nicolas Sarkozy suggested that the dogma of economic growth is no longer sustainable; and proposed replacing Gross Domestic Product (GDP) with a new measure that takes account of the quality of living, e.g., joie de vivre index.  Upon Sarkozy’s request, the Commission on Measurement of Economic Performance and Social Progress (CMEPSP) was created, aiming to identify the limits of GDP as an indicator of economic performance and social progress and to suggest more relevant indicators of social progress (Stiglitz, Sen, and Fitoussi 2009).  In the report by the Commission, Stiglitz and Sen write:  When there are large changes in inequality (more generally a change in income distribution) gross domestic product (GDP) or any other aggregate computed per capita may not provide an accurate assessment of the situation in which most people find themselves. If inequality increases enough relative to the increase in average per capital GDP, most people can be worse off even though average income is increasing (Stiglitz, Sen, and Fitoussi 2009).    The findings of this paper provide some confirmation of Stiglitz and Sen’s insight: if the fruit of economic growth through trade liberalization is unequally shared, making the rich richer and/or the poor poorer, the rise in average income may not lead to increase in subjective well-being of people thereby making them vote for more openness.  In fact, experimental economists have found that factors other than income growth affect individuals’ assessment of their own subjective welfare, and that these same factors may also influence individuals’ responses to economic policies (Frey and Stutzer 1999, 755-778).  The   42 findings of this paper suggest that widening income inequality is one such factor in the politics of trade liberalization.  The findings thus have an important policy implication: Policymakers committed to trade liberalization should take into account the issue of how the policy influences those at the bottom or at the top of the income distribution; as well as the issue of whether it increases average income of the country.  The findings also suggest that the gap between information contained in aggregate GDP data and what accounts for common people’s subjective well-being may not stem from ignorance or irrationally of individuals; rather, the gap may be better accounted for by complexity of the well-being of individuals.  The facts that differences in perceived job prospects are more important determinants of support for trade than economic literacy, and that differences in perceived job prospects here have psychological roots – given that participants in political science subject pool are undergraduate students that can be viewed as constituting potentially a socio-economically homogenous group – uncover limitations of both material self-interest based explanations and Hainmueller and Hiscox’s or Caplan’s economic idea based explanations of trade preferences.   The purpose of this paper was to provide an empirical test of how much perceived inequality and poverty are tolerated, as a concomitant of income growth generated by trade liberalization, using undergraduate students in a Canadian university.  Further study is needed to see if the findings of this paper can be generalizable to countries with different socio-economic development levels or socio-economic values.  In fact, experimental economists find that empirically income increases subjective well-being at low levels of development but once a threshold around $10,000 is reached, the average income level in a country has little effect on average subjective well-being of the people (Frey and Stutzer 2002,   43 402-435).  Also, Alesina et al. find that there is a large negative significant effect of inequality on subjective well-being of the people in Europe but not in the United States (Alesina, Glaeser, and Sacerdote 2001, 187-254).  Given these existing findings, equality-growth trade- offs may systematically differ across countries with different development levels or a given country’s dominant social, economic values.  Policymakers committed to trade liberalization may need to get a better sense of the country-specific level of equality-growth trade-offs so that they can incorporate it into their calculation when designing and implementing trade policies aimed at improving people’s well-being and thereby gaining more public support.        44 2 THE SOURCE OF FEMALE PROTECTIONISM: THE CONSEQUENCE OF KNOWLEDGE GAP IN ECONOMICS?  2.1 Introduction  A growing body of scholarly research has examined the determinants of public attitudes toward trade liberalization.  The survey-based studies have largely agreed on the findings that protectionist backlash is highest among respondents with the lowest levels of education, and that it also has a strong gender orientation.  The positive relationship between education and support for trade liberalization has been interpreted as support for economic trade theory, i.e., the Heckscher-Ohlin Stopler-Samuelson theorem, suggesting that trade preferences are primarily a product of distributional concerns derived from the labour market implications of trade.  Here, educational attainment simply serves as a proxy for skill level.  Recently however, a group of scholars has argued that the relationship between education and trade attitudes may reflect not so much distributional concerns, which differ with respect to skill level, but rather education effects per se; that is, exposure to economic ideas at university makes the primary difference in attitudes toward trade liberalization (Caplan 2002, 433-458; Hainmueller and Hiscox 2006, 469-498; Caplan 2006, 367-381; Hainmueller and Hiscox 2007, 399-442).  In a similar vein, recent studies that have addressed the source of the gender gap in protectionist sentiment have argued that differences in university-level education experience are the most plausible explanation for the puzzle of female protectionism and have done so by presenting data where the gender effect interacts with education (Caplan 2006, 367-381; Burgoon and Hiscox 2003; Burgoon and Hiscox 2008).  This argument appears to have been well-received by mainstream economists who largely agree that it is   45 economic knowledge rather than ideology that trade is good (Burgoon and Hiscox 2008; Caplan 2007).   This paper investigates the sources of female protectionism by testing the claim that exposure to economic ideas at university indeed accounts for the gender gap in protectionist sentiment.  The gender difference in policy preferences is in fact not a phenomenon that is confined to the case of trade policies.  It has been found that gender makes a difference not just when it comes to trade policies, but also other policy areas as well, including government spending, gun control, and some aspects of social welfare policy (Sapiro 2003, 601-634).  As I will detail in later sections, even among policy questions formulated as trade issues, gender gaps vary in scope depending on the nature of the question.  Indeed, the controversy over trade liberalization has many different facets.  The theory of comparative advantage explains how trade can create benefits for both parties, even when one party can produce goods more efficiently than the other; and mainstream economists, who are advocates of the theory, usually evaluate economic policies based on the efficiency-enhancing effects of the given policy or aggregate total benefits to be accrued by the policy.  When it comes to public preferences, however, efficiency or growth is hardly likely to be the only criteria to determine their preferences for trade policies.  In an attempt to provide a more plausible explanation for female protectionism, I explore the patterns of the gender gap that are found to vary depending on which aspect of trade liberalization is primed in the survey question, whereby I formulate alternative hypotheses.  I then use survey experiments to directly test these hypotheses along with Burgoon and Hiscox’s claim that the gender gap stems from differences in exposure to economic ideas at university.    46  This main argument of this chapter is twofold.  First, I demonstrate that the gender gap is not simply reducible to differences in university-level education experiences.  Contrary to Burgoon and Hiscox’s hypothesis, this paper finds that the gender gap remains strong even after exposure to economic ideas is controlled for.  Admittedly, education matters: the results of the experiments support that male university students are more likely than female students to study economics or a related field, and that there is a clear difference in protectionist sentiments between those who have completed economics courses and those who have not.  The results, however, also show that the gender gap still remains strong, even when only those who have completed economics courses are taken into consideration. These findings indicate that even when exposed to the same economic ideas, such as the theory of comparative advantage, female students may not accept the ideas to the extent that male students do, probably for the same reasons that female students prefer studying economics less than male students do, on average.  Second, I further demonstrate that the gender gap may be best explained by gendered responses to those facing hardships.  The survey experiments find evidence that women tend to identify as being significantly more “sympathetic” than men; and that when the gender difference in this personality trait is controlled for, the gender effect significantly disappears.  In fact, public opinion scholars have long suggested that citizens organize their policy opinions around visible social groupings.  They have argued that public opinion is shaped not only by material interests7 that people see at stake in issues, or symbolic  7 In fact, numerous studies on public opinion have found that self-interest often plays surprisingly little or no role in determining policy preferences.  For example, self-interest fails to influence mass preferences in  policy issues such as bussing, health insurance, unemployment programs, the Vietnam War, and affirmative action (Sears et al. 1980, 670-684; Lau and Sears 1981, 279-302; Kluegel and Smith 1982, 518-532; Kinder 1986, 151- 171).   47 predispositions, such as ideology or party identification, but that it is also guided in powerful ways by the sympathies and resentments that people feel toward the social groups they see as the principal beneficiaries or victims of the policy (Nelson and Kinder 1996, 1055-1078; Conover 1988, 51; Sniderman, Brody, and Tetlock 1991).  Drawing on the insights of public opinion scholars, to which surprisingly little scholarly attention has been paid in the analysis of trade preferences, as well as  the findings of experimental economists that women are more sensitive to social cues (for review, see Croson and Gneezy 2009, 448-474), I test and confirm the hypothesis that it may be gender differences, specifically the degree of sympathy for those social groups implicated in the policy, that generate the gender gap in protectionist sentiment.  This paper proceeds as follows.  In the first section, I briefly review the recent literature on female protectionism.  Here, I first discuss the Burgoon and Hiscox hypothesis that differences among men and women in exposure to economic ideas and information may be generating the gender gap in attitudes toward trade.  By drawing on the finding of both the public opinion studies that policy preferences are often a function of sociotropic judgments, and experimental economics about gender differences in social (other-regarding) preferences, I propose alternative hypotheses.  In the second section, the proposed hypotheses along with the Burgoon and Hiscox hypothesis are tested: the experimental design is introduced and the results are presented.  In the last section, I conclude.    48 2.2 The gender gap in protectionist sentiments and sociotropic hypotheses  Numerous empirical studies have shown that women are more likely than men to be protectionist, and that this gender gap remains even after individuals’ socio-economic characteristics are controlled for.  Shapiro and Mehajan (1986), in their paper examining gender differences in policy choices in the United States, found that gender differences in opinion toward regulation-protection policies were pronounced and had become more salient over past two decades from the 1960s to 1980s.  Gidengil (1995) also found a sizable gender gap in Canadians’ support for the United States-Canada Free Trade Agreement in the 1988 Canadian federal election.  She pointed out that less than half (44%) of Canadian women supported the proposed trade agreement by the end of the election campaign, whereas 60% of Canadian men did.  More recently, the literature examining mass support for trade liberalization has also found a statistically significant gender gap, namely, that males are more likely than females to support trade liberalization by approximately eight to ten percentage points (Caplan 2002, 433-458; O'Rourke et al. 2001, 157-206; Caplan 2006, 367- 381; Baker 2003, 423-455; Baker 2005, 924; Beaulieu 2002, 99-131; Mayda and Rodrik 2005, 1393-1430; Beaulieu, Benarroch, and Gaisford 2004, 113-136).   Among the studies that have directly addressed this puzzle, Burgoon and Hiscox’s recent work merits attention (Burgoon and Hiscox 2003).  Burgoon and Hiscox argue that differences in education experience – i.e., exposure to economic ideas at university – are the most plausible explanation for the gender gap by presenting data where the gender effect interacts with education.  To support their argument, they highlight two main findings: (1)   49 the gender gap is most pronounced between university-educated men and women (Burgoon and Hiscox 2003); and (2) the gender gap increases with age, which they argue reflects the fact that decades ago, female university  students  were much less likely to study economics than they they are now.  In their subsequent paper, Burgoon and Hiscox test the most plausible explanations for the gender gap -  the distributional effects of trade suggested by trade theories, the sensitivity to income risks associated with maternity (compassion for the less fortunate), other political values such as party affiliations, and economic ideas and information -  and confirm that it is economic literacy that generates the strong gender orientation in attitudes toward trade (Burgoon and Hiscox 2008).  Caplan (2007) also tests Burgoon and Hiscox’s hypothesis by employing a new dataset that includes a survey of economists as well as the general public (Caplan 2007).  He shows that the interaction effect between gender and education (university attendance) is robust – i.e., that the gender gap is driven by disagreement mostly between university-educated men and women, and that the least educated men and women, on average, agree.  Although Caplan finds no evidence of the age effect – i.e., the decline in the gender-education effect over time – which suggests that men and women probably have different levels of interest in economics in the first place, he concludes that his findings are basically supportive of Burgoon and Hiscox: that women are more protectionist than men because women know less about economics than men.  This paper tests this claim by challenging the idea that female protectionism simply reflects women’s ignorance of economic knowledge, or their irrationality – “irrational” in the sense that they are more likely to vote for protectionism even though they prefer the actual effects of trade liberalization (Caplan 2007).  There is in fact a widespread consensus among economists that trade liberalization is good for national economic welfare.  There is little   50 doubt that trade liberalization, at least in theory, is an important engine of growth for countries at different stages of development by contributing to a more efficient allocation of resources within and across countries.  Accordingly, the prevailing protectionism among the public has often been considered a mere reflection of the public’s ignorance or irrationality (Caplan 2007; Irwin 2005a).  There are, however, at least two good reasons to believe that the argument that the gender gap is a product of differences in exposure to economic ideas is flawed.  First of all, the hypothesis fails to explain why the gender gap in protectionist sentiment is not usually found in less developed countries (LDCs), while it is robust in most developed countries.  Trade theory has been universally dominated by the theory of comparative advantage – i.e., the win-win theory of trade – and there is little reason to believe that the gender difference in education is less pronounced in LDCs.  If differences in exposure to economic ideas and information account for the gender gap, we should see a consistent gap in support of trade liberalization across all countries.  Empirically however, this is not the case.  Although most of the scholarly work on the gender gap has thus far focused on developed countries, Beaulieu and Napier (2008) in their analysis employing National Identity surveys from 1995 to 2003, include LDCs, where they find no gender differences.  They show that the level of support for trade liberalization among both men and women in LDCs is as low as it is for their counterparts in developed countries, which Burgoon and Hiscox’s hypothesis does not account for (Beaulieu and Napier 2008).  Second, while Burgoon and Hiscox refer to their explanation as an “ideational” account, and thus distinguish it from the Heckscher-Ohlin based explanations, both accounts clearly share one core idea: that people give material interests far more credit than they do any other values in life, though one is more about private material interests and the   51 other about Gross Domestic Product (GDP).  Economists evaluate economic policies or their outcomes by asking whether they make people better-off. They also assume one’s well- being depends on her or his material well-being – i.e., that one can improve well-being by increasing his or her own income;- and that public policy aimed at increasing the income of the nation as a whole leads to greater well-being of all.  I argue, however, that it is too naïve to assume that individuals place so much emphasis on GDP as economists do once they as well are exposed to economic ideas.  What seems missing from this line of thought is that trade liberalization is not “socially” costless, and that it is by nature a “political issue” – i.e., an issue about allocation of resources.  Increased trade liberalization has redistributive consequences that generate new economic winners and losers.  Given that individuals have the ability to externalize the social costs, the distributive effects of trade liberalization may influence individuals’ perception of free trade – not only  their perceptions of those directly affected but, also their perceptions of those relatively unaffected (neutrally affected) by free trade.  That is, it could be that people are not inclined to support free trade because they are not “social morons,” as which Sen refers to those propelled entirely by their material self- interest (Sen 1990: 336).  In fact, the public opinion literature has suggested that it may not be exceptional that individuals who will either benefit from free trade or be neutrally affected might take a strong protectionist position.  Although it is widely assumed that individuals basically adopt policy preferences that further their private interests, numerous studies on public opinion have provided strong evidence that immediate and tangible material self-interest has only a minimal influence over policy preferences (Sears et al. 1980, 670-684; Sears and Funk 1990, 247-271; Wolpert and Gimpel 1998, 241-262).  The empirical evidence has supported the so-   52 called “sociotropic hypothesis” that individuals do not pay so much attention to their own problems and achievements in the formation of their attitudes about political issues, but rather to the problems and achievements of a larger social “group” (Kinder and Kiewiet 1981, 129-161; Mutz 1998; Mansfield and Mutz 2009b, 425-457).  One might argue that Burgoon and Hiscox’s ideational account corresponds to the sociotropic hypothesis in the sense that they also predict that people’s perceptions of collective conditions – i.e., the effect of trade on the “national” economy, which the comparative advantage theory tells us is always positive – reliably influence their trade attitudes.  It should be noted, however, that the sociotropic hypothesis does not postulate the effect of simple national (or in-group) interests.  While the ideational account is mostly related to people’s own interests or national (in-group) interests as a whole, psychological evidence tells us that sympathy for others (the out-group) should also influence individuals’ policy evaluations.  According to the sociotropic hypothesis, a “group” is defined as “any set of people who can constitute a psychological entity for any individual” (Adorno et al. 1950, 123-196; Kinder 1998).  As Kinder indicates, groups under this definition do not require formal membership or interpersonal contact among members.  Distinction between in-groups and out-groups will thus differ in accordance with the issue being considered (or how that issue is framed).  For example, when considering international issues, nationality should be the criterion to distinguish between in- and out-groups, and for affirmative action, race should be the criterion.  In this regard, groups are not fixed identities, and they do not dictate preferences. What “groups” do in the sociotropic hypothesis is change the utility functions of individuals by widening the scope of self-interest to that of their in-group, or by including their feelings toward the out-groups whom they see as the principal beneficiaries or victims of the policy.    53 In fact, as Mutz reviews in her book, Impersonal Influence, the literature on American public opinion studies is replete with evidence that people’s attitudes toward politics are influenced by their perceptions of others8’ experiences (Mutz 1998; Mutz 1992, 89-122). Jacobs and Shapiro, for example, find that there is an interesting disjuncture between the public’s overall contentment concerning their personal healthcare and the public’s dissatisfaction with the quality of healthcare available to others: while a stable 84-89% of Americans report being personally satisfied with the quality of the healthcare they receive from doctors, only half as many agree that other patients enjoy high-quality treatment (Mutz 1998; Jacobs and Shapiro 1995, 411).  This suggests that citizens may simplify policy issues that are often complex and multifaceted by turning them into judgments and feelings about visible social grouping, and public opinion may be a simple reflection of citizens’ political thinking about the groups involved in a given policy.  The process of public opinion formation around economic issues, such as trade liberalization, does not seem to be an exception to such sociotropic influences.  By drawing on Schelling’s discussion (1984) that political process tends to be more favourable toward those groups whose characteristics are known over those groups or people in general whose characteristics are less known, Ann Kruger (1989), for example, argues that such “identity bias” is likely to strengthen protectionist sentiments.  That is, the knowledge of the losers’ (i.e., losers from free trade) identities evokes sympathetic attitudes toward their plight, which likely leads people to favour more protection.  In this paper, I argue that female protectionism is in large part explained by the gender difference in social preferences – e.g., sympathy for the groups implicated in the  8 According to Mutz, “others” in this case refers to anonymous mass collectives beyond their personal contacts (Mutz 1998).   54 policy issue in particular.  In fact, the idea that women and men have different social values and concerns is not new: scholars have argued that gender differences in political attitudes are captured in the notions of “social woman” and “economic man” (Gilligan 1982; Welch and Hibbing 1992, 197-213; Gilens 1988, 19-49; Gidengil 1995, 384).  Gidengil for example finds that corresponding to the image of “social woman,” women are much more sceptical of the virtues of competition and substantially more egalitarian than men, which consequently shapes opinion on the Canada-United States Free Trade Agreement (Gidengil 1995, 384).  She also suggests that women and men may differ in the concerns they bring to bear in evaluating policies.  Experimental economists and psychologists have also studied social preferences of the genders.  A number of studies find evidence that women’s social preferences – e.g.., other-regarding behaviour in the form of altruism and inequality-aversion – are different from men (Bolton and Ockenfels 2000, 166-193; Fehr and Schmidt 1999a, 817-868; Becker 1971; Andreoni 1990, 464-477).  But it appears that the findings of these studies are varied: for example, some ultimatum and dictator game studies find that women are more trusting than men, but others find that the reverse is the case (See Croson and Gneezy 2009 for review).  After reviewing the findings on gender differences provided by experimental economists and psychologists, Croson and Gneezy conclude that women are not more socially oriented, but that their preferences are more situationally specific and malleable than those of men (Croson and Gneezy 2009, 448-474).  This suggests that gender differences may be interpreted as women simply being more responsive to the conditions of experiments.  In short, it is not yet clear whether gender differences lie in their differential fundamental values or sensitivities to social cues, which is also tested in the following sections.  By drawing upon these findings from the public opinion studies and experimental   55 economics noted above, specific hypotheses are presented and tested in the following sections.  2.3 Explaining the gender gap in protectionist sentiment  2.3.1 Preview and hypotheses  Before I begin a detailed analysis of female protectionism through lab experiments, I first briefly examine the survey data of Americans conducted by Chicago Council on Foreign Relations (CCFR) in 2004.  The CCFR is an ideal dataset for getting an approximate idea of which facet of trade liberalization may generate the gender gap because it contains a number of survey questions about trade liberalization, including questions about the North American Free Trade Agreement, agricultural subsidies, and outsourcing.  Table 2.1 lists the CCFR’s seven main questions related to trade liberalization.   Table 2.1: Survey questions related to trade in CCFR 2004 and gender differences Variable Question Protectionism Globalization in general Do you believe that globalization, especially the increasing connections of our economy with others around the world, is mostly good or mostly bad for the United States?  - Mostly good - Mostly bad   M = F Outsourcing Currently there is a debate about outsourcing US jobs; that is, moving jobs to countries where wages are lower.   Which position is closer to yours?  - Outsourcing is mostly a good thing because it results in lower prices in the US which helps stimulate the economy and create new jobs - Outsourcing is mostly a bad thing because American workers lose their jobs to people in other countries  M < F   56 Variable Question Protectionism Trade liberalization with assistant programs Which of the following three positions comes closest to your point of view about lowering trade barriers such as tariffs?  - I favour agreements to lower trade barriers provided the government has programs to help workers who lose their jobs - I favour agreements to Lower trade barriers, but I oppose government Programs to help workers who lose their jobs - I oppose agreements to lower trade barriers  M = F Free Trade Agreement of the Americas The U.S. and most countries in North, Central and South America have been discussing the possibility of having a Free Trade Agreement of the Americas similar to what the U.S. now has with Mexico and Canada in NAFTA, do you favour or oppose this idea?  - Favour - Oppose  M = F International trade in general Overall, do you think international trade is good or bad for  - The U.S. (United States) economy - American companies - Consumers like you - Creating jobs in the U.S. (United States) - The environment - Job security for American workers - Your own standard of living  M = F NAFTA  Overall, do you think the North American Free Trade Agreement, also known as NAFTA, is good or bad for  - the U.S. (United States) economy - American companies - Consumers like you - The Mexican Economy - Creating jobs in the U.S. (United States) - The environment - Job security for American workers - Your own standard of living - Creating jobs in Mexico    M < F (The Mexican Economy and Creating jobs in Mexico)  M = F (The rest) Agricultural subsidies Do you favour or oppose the US government giving subsidies to small farmers, who work farms less than 500 acres?  - Favour - Oppose  M < F Source: Global Views 2004: American Public Opinion and Foreign Policy, Chicago Council of Foreign Relations.     57 As shown in Table 2.1, the gender gap varies depending on the question.  When it comes to questions about globalization in the abstract – e.g., increasing economic connections – and even those asking opinions about “new” free trade agreements encompassing all of the Americas, no significant gender difference is found.  Women are no less positive than men in their assessment of the effects of international trade on the national economy, companies, consumers, job creation in the U.S., the environment, job security for American workers, and standard of living.  On the other hand, when it comes to questions about specific trade policies – especially those that contain information about specific social groups implicated in the policy debates – the gender gap is more pronounced.  Women are less favourable toward outsourcing, something that hints at possibly taking jobs from American workers, and are more supportive than men of agricultural subsidies. Moreover, the differences are statistically significant.  The most interesting finding is that women tend to evaluate the effects of NAFTA on the Mexican side more negatively than men: women appear to be more sceptical of NAFTA being good for the Mexican economy and Mexican jobs.  Yet there is no gender difference found in assessing NAFTA in terms of its effect on American economy, American companies, consumers, American jobs, or respondents’ standard of living.   What, then, precisely explains the gender gap?  Which facets of trade might trigger gendered responses?  In this paper, I test the following hypotheses:  H1: (Burgoon and Hiscox’s hypothesis) The differences between men and women in economic knowledge may create the gender divide over trade related issues.  H2: A female preference for greater protection may be a function of females’ greater sympathy for those facing hardships.    58 H3: A female preference for greater protection may be a function of females’ greater degree of inequality-aversion.  H4: A male preference for freer trade may be a function of males’ higher propensity for pro- competition and pro-efficiency.  H5: The gender gap may stem not so much from deep-rooted value differences as females’ being malleable/susceptible to issue frames.   2.3.2 The survey experiment  This paper uses a survey experiment to directly test the hypotheses above. While a number of studies have been conducted to interpret differences in the political preferences of women and men, most of these studies have used data from available public opinion surveys (Burgoon and Hiscox 2003; Burgoon and Hiscox 2008; Welch and Hibbing 1992, 197-213; Gilens 1988, 19-49; Conover 1988, 985-1010; Cook and Wilcox 1991, 1111-1122).  The method of the survey experiments suits better for testing the given hypotheses than the public opinion survey methods for the following two reasons.  First, the conclusions of the survey-based empirical studies were drawn mostly from inference rather than direct evidence. The method of the survey experiments has merit in this regard, as it enables us to directly test the hypotheses.  Second, it is often argued that the experimental method helps internal validity at the considerable expense of external validity – i.e., the extent to which the results of a study is generalized to a general population.  Given the findings that the gender difference in protectionist sentiment largely stems from differences in university-level education experience, however, the undergraduate sample can enhance internal validity with little loss of external validity.    59 The experiment was conducted in December 2010 at the Public Opinion Laboratory at the University of British Columbia.  A total of 190 undergraduate students from the political science subject pool participated in the study.  The experiments conducted in the lab are computer-based tests designed to study effects of different treatments on responses.  For this experiment, subjects were randomly allocated to four groups, with each group receiving different introductions to the survey question about agricultural subsidies.  We used the question about agricultural subsidies to measure protectionist sentiments as it is one of the CCFR survey questions where the gender gap is most pronounced.  Among the four groups, one group received no introduction to the survey question about agricultural subsidies (control group), while the other groups received different introductions that describe the pros and cons of the subsidies – i.e., one emphasizing the economic inefficiency of the subsidies and the other emphasizing the subsidies’ income protection effect for farmers. The exact wordings are shown below:  Group 1: No introduction  Group 2: Hardship introduction  Some argue that government should provide subsidies to farmers because otherwise their earnings fluctuate drastically.  They say that subsidies even out farm income and prevent hardship.  Group 3: Efficiency introduction  Some argue that government should eliminate subsidies for farmers because workers displaced from agriculture will get new jobs in other industries.  They say that subsidies waste resources by allowing uncompetitive farms to remain in business.  Group 4: Both introductions (the order of the two introductions was rotated)  Some [Others] argue that government should provide subsidies to farmers because otherwise their earnings fluctuate drastically.  They say that subsidies even out farm income and prevent hardship.  Some [Others] argue that government should eliminate subsidies for farmers because workers displaced from agriculture will get   60 new jobs in other industries.  They say that subsidies waste resources by allowing uncompetitive farms to remain in business.    After the frames were given, all subjects were asked the same core question about their attitudes toward agricultural subsidies:  Do you feel that the Canadian government should provide subsidies to farmers? Yes No   The specific claims in the frames that were provided to the different groups were designed to prime either subsidies’ efficiency-reducing effect or their risk-reducing effects for farmers, or both.  If the notion of “economic men” and “social women” does account for the gender gap in protectionist sentiment, the different frames will produce a gendered response: men will be more responsive to the Economic efficiency frame, whereas women will be more responsive to the Hardship frame.  If the notion above is confirmed, then we will find that the gender difference will be most pronounced in Group 4, which received both frames, because the combined frames likely lead men and women in different directions, thereby polarizing them even further.  In addition, I included questions to measure subjects’ economic literacy to test Burgoon and Hiscox’s hypothesis (H1).  All of the subjects were asked to answer these questions at the end of the experiment regardless of which group they were in.  First, I asked subjects if they had taken any economic courses.  Then, I used a simple quiz to gauge whether subjects knew about the theory of comparative advantage and did indeed understand the theory.  The exact wordings for these questions are below:     61 Have you completed a university course in economics? Yes No Don’t know Refused  The central concept in economic theory about international trade is:9 Double effect Comparative advantage Scarcity Rational expectations Don’t know Refuse  According to economic theory, which countries benefit from international trade? All countries Small countries Poor countries Wealthy countries Don’t know Refuse  If Burgoon and Hiscox’s hypothesis is correct, we will observe clear gender differences in answers to these questions, but no differences when only those who are economically literate are counted.  The results of the experiment are presented in the following section.   2.3.3 Results: Economic literacy and the gender gap  Table 2.2 reports the simple frequencies of each answer to the subsidy question by gender. As expected, a clear gender difference is found in opinions about agricultural subsidies. While a total of 77% of the female subjects supported agricultural subsidies, only 60% of the male subjects did.    9 Correct responses are shown in bold.   62 Table 2.2: Gender and attitudes toward agricultural subsidies Question: Do you feel that the Canadian government should provide subsidies to farmers?  Female (N = 95) Male (N = 75) Difference Yes 73 (76.8%) 45 (60.0%) 16.8% (M < F) Notes: The sample size is 170, after excluding those who answered “Don’t know” or refused to answer.   To see if the difference in exposure to economic ideas or economic literacy does account for the gender gap, I first ran three-way tables with the three variables – (1) gender and (2) economic literacy as independent variables and (3) subsidies as a dependent variable – and displayed them in mosaic plots.  Table 2.3 displays the three-way tables.  Table 2.3 The effect of economic literacy on the gender gap  Economic literacy: Have you completed a university course in economics? Question 1: Do you feel that the Canadian government should provide subsidies to farmers? Gender Not completed (N = 45) Completed (N = 125) Female 26 (87.7%) 47 (72.3%) Yes (to subsidies) Male 11 (73.3%) 37 (82.3%) 34 (56.7%) 81 (64.8%)  Economic literacy: The central concept in economic theory about international trade is Comparative advantage. Question 2: Do you feel that the Canadian government should provide subsidies to farmers? Gender Incorrect answer (N = 27) Correct answer (N = 124) Female 11 (72.3%) 51 (76.1%) Yes (to subsidies) Male  7 (58.3%) 18 (66.6%) 32 (56.1%) 83 (66.9%)  Economic literacy: According to economic theory, which countries benefit from international trade? All countries Question 3: Do you feel that the Canadian government should provide subsidies to farmers? Gender Incorrect answer (N = 56) Correct answer (N = 104) Female 31 (83.8%) 38 (71.7%) Yes (to subsidies) Male 13 (68.4%) 44 (78.6%) 29 (56.9%) 67 (64.4%)     63   The first table provides evidence of the strong impact of education experience on attitudes toward subsidies: subjects who have completed a university course in economics in general tend to be less supportive of subsidies than those who have not.  While 82.3% of those answered that they had not completed a university course in economics supported subsidies, whereas only 64.8% of those who answered they had completed a course were pro-subsidies.  The gender orientation, however, remained large even after education experiences were controlled for: even when only those who have completed an economics course were taken into account, it was found that about 15% more females are supportive of subsidies than males (72.3% versus 56.7%).  The third table displays similar outcomes.  In general, those who answered that “all countries” benefit from international trade were less likely to support subsidies than were those who chose “wealthy countries” or the other options in the question.  64.4% of those who got the answer right were pro-subsidies, whereas only 78.6% of those who got the answer wrong were so.  The gender gap, however, did not become any narrower even when only those who got the answer right were considered: about 15% more females were supportive of subsides than males in both groups (both the right answer group and the wrong answer group).   The second table displays somewhat different outcomes.  While the gender effect is clear here as well, it appears that the level of economic literacy, measured by the question of whether he or she knows of the concept “comparative advantage,” had no influence on subjects’ preferences for subsidies.  This outcome is interesting, given that the question asking subjects what “the central concept in economic theory about international trade” is enables us to obtain probably the most objective indicator to measure economic literacy among all three questions here.  For example, it is hard to conclude that the difference in   64 preferences for subsidies between those who have completed an economics course and those who have not (the first literacy question) is reducible simply to the economic literacy, especially considering that students who are interested in taking an economics course likely differ in the first place from those who have little interest.  Those who are interested in taking courses in economics might well be more pro-market, pro-competition, and pro-trade than those who are not, even before they become exposed to economic ideas and information.  Likewise, the third question asking subjects to answer which countries benefit from international trade (according to economic theory) appears not to free itself from its ideological components.  As the question did not specify the term “economic theory,” it might well have served as an ideology question measuring whether she or he buys the theory of comparative advantage or not, as well as an indicator to measure the level of economic literacy.   Despite the slight differences depending on different literacy questions, the implication of the results above is clear: taken as a whole, the results suggest that the strong gender orientation is not likely to be reducible to gender differences in exposure to economic ideas at university.  The results are in line with Hainmueller and Hiscox (2006) in the sense that both suggest that the effects of education on individual trade preferences may be not so much a product of distributional concerns linked to job skills, as it is the effects of “ideas.” The results, however, do not confirm Burgoon and Hiscox (2003): the differences in educational experience do not appear to be generating the gender gap in protectionist sentiments.  While it appears to be true that women are less likely than men to complete a course in economics at university, the gender gap still remains large even after education   65 experiences are controlled for.  This suggests that the differences in education experience do not fully account for the gender gap in attitudes toward subsides.   If not education experiences, then why might women view subsidies less favourably than men?  The alternative hypotheses (H2, H3, and H4) were also tested.  To measure the personality trait “sympathy,” I used a self-report of “sympathy” from our personality battery. The original variable has a scale from 1 to 10, where 1 means “most sympathetic” and 10 means “least sympathetic”; and I rescaled the variable to have a scale from 1 to 4, where 1 means “least sympathetic” and 4 means “most sympathetic.”  To measure inequality-averse attitudes, I used answers to a question asking how much respondents think should be done to reduce the gap between the rich and the poor in Canada on a scale from 1 to 5, where 1 means “much less” to 5 “much more”.  To measure respondents’ market orientation, I used answers to the question asking how much respondents agree with the statement that “the government should leave it entirely to the private economy to create jobs” with a scale from 1 to 5, where 1 means “strongly disagree” and 5 “strongly agree”.     66  Table 2.4 Individual support for agricultural subsidies – ideology, inequality-aversion, sympathy, and the gender gap (Full sample) Dependent variable = 1 if respondent favours agricultural subsidies to farmers (= 0 if opposes).  Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Female  0.168* (0.071)  0.153* (0.072)  0.146* (0.073)  0.145* (0.073)  0.151* (0.075)  0.125 (0.077) Economic education  -0.177. (0.093) -0.168. (0.093) -0.164. (0.095) -0.135 (0.096) -0.175. (0.100) Ideology    -0.027 (0.019) -0.026 (0.023) -0.025 (0.020) -0.023 (0.021) Inequality- aversion     0.010 (0.044)  0.007 (0.047) -0.017 (0.050) Market- orientation     -.0.029 (0.049) -0.018 (0.052) Sympathy       0.115*** (0.034)  Log likelihood -101.89 -99.90 -98.84 -98.81 -96.81 -91.12 Observations  170  170 170 170 167 167 Notes: The numbers shown in the table are marginal effects XXYP  /))|1( with standard errors in parenthesis. Significant codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1    Up to this point, I have just been examining response frequencies for different categories of individuals.  To gauge the impact of the variables above on the gender gap in a more precise manner, I estimated individual preferences for subsidies using respondents’ answers to the question about whether they favoured or opposed giving subsidies to farmers as the dependent variable, and included the variables suggested above in the models.  Table 2.4 reports the results from logit estimations of a series of models that include a number of explanatory variables suggested above.  Consistent with the findings above, the estimated gender effect is to increase support for subsidies by 17% (s.e. 7%); and education experience (whether she or he has completed a course in economics or not) reduced support for subsidies by approximately 18% (s.e. 9%).  Left-right ideology, market-orientation, and inequality-aversion had no impact on responses.  But the variable “sympathy” had a significant impact on responses; moreover, when sympathy was included, we observed that   67 the gender effect became insignificant.  I find evidence that women tend to identify as being significantly more “sympathetic” than men – the mean difference in sympathy between men and women on a scale from 1 (most sympathetic) to 10 (least sympathetic) is 0.59, and the difference is statistically significant (p=0.019), and that the gender effect is largely explained by the personality trait, “sympathy.”   2.3.4 Results: Economic man and social woman?  The first table of Table 2.5 reports the simple frequencies of each type of response in each of the four experimental groups.  No introductions other than hardship were found to have significant effects on subsidy responses.  The effect of the hardship introduction, however, was the opposite of what I hypothesized: those who received an introduction about “hardship” (Group 2) turned out to be significantly more likely to be against subsidies than those in any other group.  While only 28% of respondents answered that they opposed subsidies in Group 1 (Control group with no introduction), 44% of those who received the hardship introduction answered that they did.  More interestingly, as shown in the second table of Table 2.5, this unexpected effect of the hardship introduction is observed only among men (not among women).  In fact, among the women who received the hardship introduction, slightly more answered that they favoured subsidies than was the case with women who received no frame (71% versus 79%); but the reverse was true for men.  The hardship introduction reduced men’s support for subsidies drastically: while 74% of men who received no introduction answered that they favoured subsidies, only 32% of men who received the hardship introduction answered that they did.    68  Table 2.5 Gender and the treatment effects   Question: Do you feel that the Canadian government should provide subsidies to farmers? Introduction Subsidies Subsidies to farmers All respondents (N = 170) Yes 128 (75%) Control (Group 1: N = 47) Yes 34 (72%) Hardship introduction (Group2: N = 46) Yes 26 (56%) Efficiency introduction (Group 3: N = 43) Yes 33 (77%) Both introductions (Group 4: N = 34) Yes 35 (73%)   Question: Do you feel that the Canadian government should provide subsidies to farmers? Gender Introduction Subsidies Female (N = 95) Male (N = 75) Differences All respondents (N = 170) Yes 73 (77%) 45 (60%) 17% Control (Group 1: N = 47) Yes 20 (71%) 14 (74%) -3% Hardship introduction (Group2: N = 46) Yes 19 (79%) 7 (32%) 47% Efficiency introduction (Group 3: N = 43) Yes 16 (80%) 17 (74%) 6% Both introductions (Group 4: N = 34) Yes 18 (78%) 7 (64%) 14%     To gauge the treatment effects and the gender effect interacted with the treatments in a more systematic manner, I ran a logistic regression with interaction terms.  The numbers in Table 2.6 are marginal effects with standard errors in parenthesis.  Model 1 shows that the all three treatments have no statistically significant effects on attitudes toward subsidies. While the hardship introduction has the biggest (negative) effect on support for subsidies among all, the effect is not statistically significant.  When interaction terms between gender and treatments are included, the effect of the hardship introduction becomes greater and statistically significant.  More specifically, the interaction effect between Female and Hardship shows that being female corresponds to a difference of 0.470 (47%) in the marginal effect   69 for the hardship introduction; and the changes in marginal effects suggest that being female reduces the hardship introduction’s negative association dramatically.  The results confirm that the gender effect is largely driven by the gender differences in responses to the hardship introduction.  Table 2.6: Logistic estimation: gender and treatment effects Dependent variable = 1 if respondent favours agricultural subsidies to farmers (= 0 if opposes).  Model 1 Model 2 Model 3 Introductions   Hardship introduction -0.148 (0.094) -0.141 (0.095) -0.850 (0.325)**   Efficiency introduction  0.049 (0.103)  0.074 (0.105) -0.095 (0.334)   Both introductions  0.013 (0.108) -0.002 (0.110) -0.277 (0.373) Gender Female   0.174 (0.074)* -0.024 (0.142) Female*Hardship    0.470 (0.202)* Female*Efficiency    0.097 (0.211)   Female*Both intro    0.177 (0.222) Log likelihood -102.17 -99.35 -96.21 Observations  170  170 170 Notes: The numbers shown in the table are marginal effects XXYP  /))|1( with standard errors in parenthesis. Significant codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1   Unfortunately, it is not entirely clear precisely what in the hardship introduction turned men (but not women) away from subsidies.  One thing that is clear is that males (but not females) responded sensitively (though adversely) to the hardship introduction, and did not like the idea that “subsidies even out farm income and prevent hardship” (of famers)10. In this regard, the results partly confirm the notion of “social woman” (in a comparative sense), although the results were basically driven not by females but by males.  It is noteworthy that the gender gap in preferences for subsidies was driven by the strong adverse  10 The second round of the experiments was conducted in April 2011 to test if the particular idea that subsidies even out farm income might have generated the adverse male effect.  The exact wording of the new hardship frame is as follows: “Some people say that government should provide these subsidies because farm earnings fluctuate drastically, and the subsidies protect small farmers and their families from hardship.”  Once the strong egalitarian idea implicit in the original frame was removed, the adverse male effect disappeared.   70 effect of the Hardship introduction on male responses.  As briefly discussed in the previous section, experimental economists often conclude that the gender gap in social preferences stems from females being more susceptible to issue frames (H5).  But the results here suggest that may not be the case – it was males who displayed a strong sensitivity to the frames (at least to the Hardship frame) in the experiments.  It seems unsubstantiated that the gender gap in protectionist sentiment is explained by a stronger sensitivity of females (than males) to the social conditions in the experiment, and that it may be true that women are more socially oriented than men, thereby leading women (men) to be more protectionist (less protectionist).  2.4 Conclusion  The findings of the study are summarized as follows.  First, Burgoon and Hiscox’s hypothesis is unsubstantiated.  The survey experiment demonstrated that even after controlling for education experience and economic literacy, the gender gap remained large. While it turned out to be true that female students were less likely to take a course in economics than male students, the gender gap in preferences for subsidies was not any less pronounced, even when only those who had completed a course in economics, the so-called “economic literate,” were taken into consideration.  These results suggest that the gender differences in protectionist sentiment may not stem from differences in exposure to economic ideas and information; rather, it seems more plausible to argue that the reasons that make women less interested in studying economics in the first place also likely account for the gender gap in preferences for subsidies.  Putting all these findings together, I conclude that the gender gap may have more to do with women’s’ pricing growth (the theory   71 of comparative advantage), than it does with the gender differences in education experiences at university.  Second, I found that women were significantly more likely than men to identify as being sympathetic, and that the difference in the degree of self-identified sympathy in large part explained the gender gap.  The gender differences in symbolic predispositions – i.e., ideology and inequality-aversion – however, did not account for the gender gap in preferences for subsidies.  Lastly, I found that the gender gap was driven largely by the strong adverse effect of the Hardship frame on males.  The results partly confirmed the notion of “economic men and social women” in the sense that we found a clear gender difference in responses to the frame that “subsidies even out farm income and prevent hardship” (of famers): females exposed to the frame leaned slightly more toward a pro- subsidy position, whereas males turned heavily against subsidies.  Also, it is noteworthy that unlike the well-known claims that women are more sensitive to issue frames than men, and that this differential sensitivity of men and women to the social conditions in the experiments results in differences in social preferences, it was males that responded sensitively (though adversely) to the frame.   Mainstream economists in general evaluate economic policies by assessing how a given policy is expected to affect the “individual economic welfare” of citizens or the economic growth of the nation.  Public opinion studies (and psychologists), however, have long suggested that an individual’s state of well-being is affected by changes in other people’s income, and thus, the distributional consequences of the policy should also be highlighted along with its effect on the economic growth of the nation.  By drawing the insights from   72 these studies, this paper hypothesized and demonstrated that the level of sympathy that individuals harbour toward the social groups that they see as the principal beneficiaries (victims) of the policy likely affects their policy preferences It also demonstrated that the gender differences in this personality trait, sympathy, likely accounted for the gender gap in preferences for subsidies.  The findings of this paper also suggest that both the differences in sympathy for those facing hardship, as well as the differences in concern females bring to bear in evaluating policy issues, may result in the differences in policy preferences. Admittedly, further study is needed to find the roots of the gender gap in trade policies, but what is clear for now is that female protectionism is not accounted for simply by women being less knowledgeable of economics than men.  More scholarly attention needs to be paid to how the gender differences in fundamental values and personality traits translate into differences in trade policy attitudes.      73 3 THE COMPROMISE OF EMBEDDED LIBERALISM AND GOVERNMENT CREDIBILITY  3.1 Introduction  Increasing domestic income inequality and economic insecurity have been of concern to government because of their political implications for the government’s future trade liberalization agenda.  Although scholars do not seem to have reached a consensus on the question of whether increasing openness has been a cause of job displacement and widening income inequality11, the simultaneity of the two developments seems sufficient to consider economic openness as the most prominent culprit12.  In fact, trade liberalization by its nature implies adjustment, the costs of which often have detrimental effects on the unskilled, at least in the short run.  This means, unlike what simple Hechscher-Ohlin trade theory suggests – that trade liberalization will relieve poverty (income inequality) in developing countries (relatively unskilled-labour-abundant countries by definition) – it could be the case that those adversely affected by trade liberalization are disproportionately poor even in developing countries.  Given the widespread perception of the tension between implementing trade liberalization and alleviating inequality and insecurity, governments committed to further trade liberalization may need to provide insurance and other transfers to compensate those adversely affected by trade.  This is one of the core assumptions of the embedded liberalism thesis.  11 Scholars such as Deininger and Squire (1998) suggest that trade liberalization has led to a pattern of growth that disproportionately benefits the rich thereby worsening income inequality (Deininger and Squire 1998, 259- 287).  Other scholars such as Dollar and Kraay (2000) however cast doubt on this assessment – they find that openness benefits the poor to the same extent that it benefits the whole economy (Dollar and Kraay 2004, F22- F49). 12 Freeman (1995), for example, argued that economic openness had significant deleterious effects on the labour market in the OECD countries, although it is not the only reason for widened income disparity (Freeman 1995, 16-21).   74  A long line of scholars, including Ruggie (1983), Cameron (1978), and Katzenstein (1985) to Rodrik (1997) and Garrett (1998), have argued that concerns about the relationship between trade liberalization and domestic economic inequality (and insecurity) are warranted (Ruggie 1983, 261-285; Cameron 1978, 1243-1261; Katzenstein 1985; Rodrik 1997; Garrett 1998).  But they argue, that does not necessarily turn the public that is exposed to more volatile labour markets against further trade liberalization.  According to this argument, trade liberalization and welfare expansion are mutually reinforcing: (1) trade liberalization is likely to increase demand on governments to cushion trade-induced insecurity and inequality through welfare state expansion; (2) welfare state expansion as compensation, in turn, could help maintain public support for trade liberalization.  That is, a new grand domestic bargain (new embedded liberalism compromise) is likely to be made: publics are asked to embrace the change and dislocation produced by trade liberalization; but governments will in return promise to protect those adversely affected by means of their social, economic policy roles (Ruggie 1983, 261-285; Ruggie 1996).  The new embedded liberalism compromise is, however, based on the premise that the public trusts its government’s desire and ability to cushion trade-induced insecurity and inequality through welfare state expansion.  Although government assistant packages – such as the Trade Adjustment Assistance (TAA) of the United States, for example – often come with trade liberalization, it is not clear whether those programs offered by government as compensation would make the new grand domestic bargain possible under any circumstances.  It may be that an increase in trade at time t – 1 has a discernible effect on welfare spending at time t ; that is, there may be a short-term causal relationship between   75 trade and welfare spending.  Even if that is the case, however, if the welfare spending does not lead to public support for trade, the openness-welfare nexus will not be sustainable.  For the openness-welfare nexus to be mutually reinforcing and sustainable, the public should be willing to compromise on trade liberalization policies in return for the promised assistance packages.  A problem of credible commitment arises here, because trade liberalization and its compensation often involve intertemporal, non-simultaneous exchanges between the government and the public.  The government attempts to invest in social welfare in the hope that its investments help build public support for openness, but compromise of any significance will not be made unless the government commitment to welfare expansion is viewed by its public as credible.  Likewise, there is not much incentive for the government to increase welfare spending unless welfare spending is seen as a viable solution to the political dilemma faced by democratically elected politicians committed to trade liberalization.  Social factors such as the public’s confidence in strong, effective governmental institutions to protect them against vagaries of free market economy, however, take long time to develop. In other words, the fear of external economic insecurity takes time to be effectively mitigated through welfare programs and social insurance, while the deleterious impacts of trade liberalization can be perceived to be a more imminent threat.  Moreover, it could also be an issue to some people that trade liberalization, once implemented, might render domestic politics impotent (Kaufman and Segura-Ubiergo 2001, 553-587; Rudra 2002, 411-445).  I argue in this paper that the conditions that make government commitments credible are one of the keys to explaining variations in public attitudes towards trade liberalization across countries.  By conditions that make commitments credible, I mean both institutions and perceptions.  The concept of credible commitment is typically viewed as a   76 matter of institutional contexts, but in this paper I focus on “perceptions” of the concept that the institutions will work – i.e., the perceived credibility of the long-term policy promise – since contextual factors such as institutions only indirectly exercise their influence on the attitudes of individuals (Jacobs 2005).  I address the following specific questions in this paper. First, I investigate how the public perceptions of the government’s desire and ability to hold onto the long-term policy promise are related to public attitudes toward trade liberalization. Second, by drawing implications from the answers to the first question, I discuss whether or not it matters to people under which type of regime they live – e.g., democratic or authoritarian regime, or social democratic welfare regime or liberal welfare regime – with respect to their acceptance of trade liberalization policies.  This paper proceeds as follows.  In the first section, I outline why conditions that make government commitments credible are essential to understanding theoretical debates on trade liberalization and welfare expansion.  Here, I briefly review and critique the literature on trade and government spending.  In the second section, I present a series of empirical models highlighting the effects of government credibility on protectionist sentiment.  Based on the findings, I conclude in the final section.  3.2 Credible commitment and individual attitudes toward trade  The embedded liberalism thesis argues that governments can build public support for trade liberalization by compensating those adversely affected by trade with welfare policies. According to the thesis, faced with increased market risks, individuals look to the government for policies that help ensure against those economic risks, which consequently   77 leads to growth of the welfare state (Rodrik 1997; Garrett 1998; Garrett and Lange 1995, 627-655).  To find the micro-foundations for the embedded liberalism thesis that link the two macro-phenomena, i.e., trade liberalization and welfare growth, scholars have studied two issues: (1) whether those who expect to be at risk after trade liberalization translate their concerns (objective economic insecurity) into subjective economic insecurity; and (2) whether subjective insecurity in fact induces changes in citizens’ policy preferences to demand protection and compensation (Garrett and Mitchell 2001, 145-177; Burgoon 2001, 509-551; Hays, Ehrlich, and Peinhardt 2005, 473; Kittel and Winner 2005, 269-293).   The argument that economic openness results in welfare growth by inducing economic insecurity among those at risk, however, tells only one side of the story.  Basically, most of the work predicting a positive relationship between openness and welfare adopts an implicit, simple “stimulus-response model of politics” to borrow Rudra and Haggard’s phrasing (Rudra and Haggard 2005, 1015-1049).  With the demand side of the trade liberalization-welfare growth nexus the point of focus, the supply side of the nexus story has been largely missing.  As Hays et al. (2005) nicely summarize, the embedded liberalism thesis is based on the idea that (1) there is universal expectation among citizens for their governments to offset the increased vulnerability and insecurity associated with trade, and to distribute the benefits of trade through government intervention or by expanding social insurance, and that (2) public support for trade liberalization depends on the willingness and ability of governments to do this successfully (Hays, Ehrlich, and Peinhardt 2005, 473). What matters from the government perspective (supply side) would be not so much the former – the issue of whether such feelings of (trade-induced) economic insecurity and demands for welfare expansion exist – as the latter – the issue of whether its commitment to welfare would help maintain or further increase public support for trade policies.  The issue   78 therefore now turns to the question of whether government commitment to welfare expansion is (and is perceived as) credible.  That is, does government have the willingness and ability to cushion trade-induced insecurity and inequality by welfare state expansion? And are they viewed by its public as credible?  By tackling these questions, this paper provides an explanation for the variation in public support for trade liberalization across countries.  Scholars exploring the dynamics behind the embedded liberalism thesis have overlooked the fact that government commitments to welfare, and individual perceptions thereof, can vary across countries.  Governments responding to globalization in different domestic institutional environments have different levels of desire and ability to commit credibly to welfare provision as compensation.  While having focused on demonstrating the (general) association between the two macro phenomena – i.e., economic openness and welfare expansion – globalization scholars have largely neglected the differences in economic and political institutions across countries.  Considering that analyses of the trade liberalization-welfare growth nexus are conducted focusing mostly on developed countries (OECD countries), it is understandable that the differences have not been the major concern to globalization scholars.  Empirically, however, government social welfare spending as a response to the similar pressure of globalization differs significantly, especially between developed and developing countries.  Spending has risen in developed countries in general, but declined albeit slightly in less developed countries (LDCs) since the 1970s, thus requiring a reassessment of the embedded liberalism thesis.  Welfare spending is not identical in various developed countries, and is in fact often marked by tremendous diversity.  Moreover, as briefly discussed above, the existing literature exploring the micro-foundations of the   79 openness-welfare nexus suggests that individuals’ perceptions of economic insecurity (and inequality) lie at the core of the embedded liberalism thesis (Rodrik 1997; Garrett 1998). Thus, for economic openness and welfare to be mutually reinforcing requires assurance that welfare will effectively mitigate individual perceptions of economic insecurity, and thereby help maintain public support for trade liberalization policies.  I therefore hypothesize that public support for trade liberalization depends on the conditions that make government commitment to welfare credible, which consequently explains the different degrees of association between openness and welfare across countries.  Problems of commitment in politics often arise from the non-simultaneity of political exchange (intertemporal bargains).  This paper begins with the observation that trade politics also has a temporal feature, which poses a commitment problem.  Ruggie (1982; 1997) suggested the idea of a “grand social bargain,” whereby all sectors of society agree to open markets, and governments, in exchange, promise to moderate the volatility of open markets and provide social safety nets and adjustment assistance as compensation (Ruggie 1983, 261-285; Ruggie 1996).  But he also seemed aware of the problem of commitment involved in the bargain, thereby emphasizing the role of the governments in enacting and sustaining the bargain.  Given the temporal feature of the grand bargain, it is not surprising that government policy initiatives to provide more welfare are sometimes met with public scepticism.  While trade liberalization is often accompanied by promises that losers from trade will be compensated, these promises are not binding.  Presumably people are likely to endorse the grand bargain only when the government promises to cushion the adverse domestic effects of open markets are viewed as credible.  If these promises are not viewed as credible, the grand bargain will not be enacted, let alone sustained.  Conditions that make such government promises credible – e.g., legacy of previously instituted welfare   80 policies or perceived effectiveness of government – vary across countries; and for instance, as Ruggie points out, LDCs have never enjoyed the privilege of cushioning their populations from vagaries of market exposure, which is likely to make security-enhancing functions of government promises not work at the same level as they do in countries with a long welfare state tradition.  In the political science literature that uses the concept of credible commitment, much attention has been paid to institutional structures that play a role in reducing political uncertainty (North and Weingast 1989, 803; North 1994).  The focus of this paper is, however, more on perceptions of the credibility of government commitment to welfare expansion than on institutions themselves, although I admit that the presence of a set of formal institutions is often considered a necessary precondition for attaining such perceptions.  It is subjective feelings of trade-induced economic insecurity that condition the levels of public support for trade liberalization; and thus, public provision of social protection matters only to the extent that the public provision does reduce those subjective feelings.  As the vast literature on bounded rationality tells us, people often have great difficulties in making inferences especially when uncertainty is involved, because of the interaction between their limited cognitive capacities and fundamental complexities of the environments they face (Simon 1991, 125-134; Jones 2003, 395-412).  Given the bounds of human cognition, what matters is not so much whether welfare expansion is indeed driven by increasing exposure to international risk generated by globalization, or whether by some other factors, such as the state of the domestic economy or partisan effects.13  It is also not  13 Kittel and Winner (2005), for example, in their paper re-assessing a study of the globalization-welfare state nexus by Garrett and Mitchell, argue that government spending is primarily driven by the state of domestic economy, and that neither partisan effects nor the international economic environment have affected public   81 so much a matter of whether welfare policies to which governments promised to devote resources are indeed designed to enhance objective social equality and insecurity14.  Rather, it is people’s perceptions that matter: insofar as people trust their governments in their willingness and ability to compensate market losers, public support for trade will be maintained, which will consequently help the social grand bargain to be enacted and sustained.  There is another reason why perceptions matter: they are not something that can be easily created or dismissed.  For example, individual perceptions of government credibility in cushioning trade-induced insecurity, which is the focus of this paper, have been shaped over a long period of time by political institutions and legacies of previously instituted public policies.  In fact, many scholars have focused on a “short-term” causal relationship between trade and government spending (Rodrik 1997; Rodrik 1998, 997-1032; Swank 2001, 133-162; Adserà and Boix 2002, 229; Mares 2004, 745-774).  They argue that governments can simply garner public support for trade by neutralizing the negative effects of market integration through social policies.  I argue, however, that if it is the case that the public perceptions of government credibility play an important role in maintaining pro-trade opinions, which will be tested in the next section, the relationship between trade and the welfare state effort should be considered not so much as a short-term relationship, but rather as a long-term and historically contingent one, as once argued by Garrett and Mitchell (Garrett and Mitchell  expenditure considerably.  In this paper, I argue, however, that the embedded liberalism thesis still holds even if Kittel and Winner’s argument is empirically demonstrated, because public support for trade depends on perceptions of welfare- or security-enhancing functions of social programs (Kittel and Winner 2005, 269-293). 14 Rudra (2005), for example, finds that formal welfare programs in LDCs were not originally designed to protect the needy; and increased welfare spending in globalizing LCDs is not redistributive – privileged groups are receiving a disproportional amount of the welfare benefits in LCDs.  In this paper, however, I emphasize more on people’s perceptions than on actual empirical findings of causation (Rudra and Haggard 2005, 1015- 1049).   82 2001, 145-177).  Scholars working with an historical institutionalist approach, an approach with which this paper falls in line, have long pointed out the importance of the analysis for history and culture in explaining different outcomes in economic performance or policy (March and Olsen 1989; Skocpol 1992; Pierson 1994; Hall and Soskice 2001).  Hall and Soskice (2001), in their work on varieties of capitalism, argue that the presence of a particular set of formal institutions is rarely sufficient to guarantee a specific outcome, and do so by indicating indeterminacy problems associated with the existence of multiple equilibria.  They thus introduce into their analysis the role of culture and historical experience that shapes “a set of shared understandings of what other actors are likely to do” (Hall and Soskice, 2001: 13).  For the purpose of this paper, the two ideas in which historical institutionalism is grounded are noteworthy: (1) a historically constructed set of institutions and policy feedbacks structure the attitudes of those involved; and related, (2) the institutions “should not be seen as entities that are created at one point in time” (Hall and Soskice 2001), 13).  In essence, I argue that previously enacted public policies (with respect to the welfare state regime under which people live) are more important than suggested policies (promises) in determining the possibility of whether or not the grand bargain is made, because the legacy of previously instituted policies shapes people’s perceptions of the credibility of the government’s commitment to its promises.  This paper thus aims to provide empirical evidence that the credibility of government promises influences welfare- and security- enhancing functions of suggested social welfare programs, thereby explaining differences in public support for trade across countries, especially between developed and less developed countries.    83 On a final note: government credibility, if it is proven that it matters, will shed new light on the literature that deals with the effects of democracy (or democratization) on welfare and trade liberalization.  It is a conventional wisdom in the existing theories15 that democracies are more sensitive than authoritarian regimes to increased vulnerability associated with trade liberalization, and thus pursue more expansive social policies to protect the vulnerable.  Moreover, recent work on the effects of regime types on trade liberalization demonstrate that democratization, meaning “a movement toward majority rule with universal suffrage in contested election” changes the optimal trade policies for leaders in the direction of further trade liberalization (Milner and Kubota 2005, 107-143; Kevin O'Rourke 2007).  Following the logic of the Heckscher-Ohlin and Stopler-Samuelson theorems, Milner and Kubota, for example, argue that in LDCs, which are by definition characterized by a lack of capital and an abundance of labour, workers and the poor tend to gain from trade liberalization.  They argue, as democratization enfranchises this new group of voters with preferences for lower levels of protectionism, more-democratic countries are more willing to open their markets.  This paper, however, casts a more sceptical eye on the potential significance of such regime types – i.e., democracies vs. non-democracies.  Admittedly, it should be noted that it is not actual trade policies but rather trade attitudes that this paper attempts to explain.  In any political system there would no perfect correspondence between public preferences on any specific issue and policy outcomes.  Nevertheless, it is hardly deniable that policy outcomes often mirror public preferences.  At least, to make sense of Milner and Kubota’s argument, it is essential to assume that workers and the poor in LDCs  15 Rudra and Haggard (2005), for example, reviewed the existing literature on the welfare state, and summarized the three theories strongly suggesting that “democracies should pursue more expansive social policies than authoritarian regimes, ceteris paribus” (Rudra and Haggard 2005, 1015-1049).  See (Meltzer and Richard 1983, 403-418; Boix 1998; Cox 1987, 508-46).   84 have preferences for lower levels of protectionism, an assumption about which I harbour suspicion.  I argue in this paper that it is government credibility that matters for offsetting economic insecurity associated with trade liberalization. If that is the case, it is hard to claim that new democracies are necessarily in a better position than their authoritarian counterparts to garner public support for trade liberalization because they are not necessarily better equipped with respect to governments’ desire and ability to provide social protection.  I argue in the following sections that what matters more for shaping public perceptions of government credibility is a legacy of previously instituted social policies – i.e., people’s subjective evaluations of the type of the welfare regime under which people live – than suggested but not yet implemented policies for potential trade losers, because such a legacy is likely to form public perceptions of government’s desire to provide social protection. Moreover, government’s ability also matters:  credibility in government’s desire to provide protection only does not suffice to garner pro-trade attitudes.  New democracies that are often characterized by weak and fragile government institutions are not any better than their authoritarian counterparts in the two respects above.  In the following section, I will empirically test whether government credibility can build support for trade.  If the public’s negative evaluation of the government’s desire and ability to provide social protection is positively associated with the protectionist sentiment of the country, this implies that at least in countries that are characterized by lack of government credibility, the embedded liberalism compromise does not serve as a viable solution to the political problems faced by the elected politicians who commit their countries to trade liberalization.   85 3.3 Empirical test: Government credibility and protectionist sentiment  3.3.1 Description of the data and empirical approach  The data I use is from the Asia-Europe survey (ASES) 2000.  The data set provides information about individuals’ attitudes toward trade.  The 18 countries surveyed are: Japan, South Korea, China, Taiwan, Singapore, Malaysia, Indonesia, Thailand, and the Philippines from East and Southeast Asia, and the United Kingdom, Ireland, France, Germany, Sweden, Italy, Spain, Portugal, and Greece from Western Europe.  In order to measure individual trade policy attitudes, I focus on survey responses to the following question: “Please tell me how much you agree or disagree with the following statement: [Country] should limit the import of foreign products.”  After deleting the “Don’t know” and “NA, refused” responses, I transformed survey questions into the dependent variable Trade Opinion on a scale from 1 to 5.  I assigned a value of 1 to respondents who answered “strongly agree” and a 5 to those who answered “strongly disagree.”  Higher values of Trade Opinion thus correspond to more pro-trade attitudes, whereas low values reflect protectionist sentiment.  I also created two binary variables, labelled Pro-trade Dummy and Against-trade Dummy.  Pro-trade Dummy is set equal to 1 for individuals opposing trade protection, and to 0 for the rest.  Against-trade Dummy is set equal to 1 for those that favour protection and to 0 for the rest.   Table 3.1 presents summary statistics of Trade Opinion, Pro-trade Dummy and Against- trade Dummy, by country.  The data set shows that on average protectionist sentiment runs high: those with protectionist sentiments (46.8%) substantially outnumber the proportions of those with pro-trade attitude (29.7%) in both regions.  There is a large variation in trade attitudes across countries.  There are countries in both regions that have publics committed   86 to trade – Singapore (64.4%) in Asia, and Germany (52.0%) and Sweden (46.4%) in Europe. Four out of five the most protectionist countries are Asian: Thailand, Malaysia, Korea and Philippines. Table 3.1 Summary data on individual attitudes toward trade (Asia-Europe Survey 2001) Trade Opinion Country Strongly Agree (1) Agree (2) Neither agree nor disagree (3) Disagree (4) Strongly disagree (5) I don’t know (6) NA, refused (7) Average Trade Opinion Pro-Trade Dummy Against-Trade Dummy Japan .046 .159 .466 .219 .045 .065 3.06 4 .282 12 .220 17 Korea .302 .350 .198 .106 .032 .013 2.21 15 .139 17 .660 4 China .137 .310 .128 .345 .073 .007 2.91 6 .421 4 .450 12 Taiwan .083 .252 .235 .331 .043 .056 3.00 5 .396 5 .355 15 Singapore .026 .158 .158 .455 .164 .039 3.60 1 .644 1 .191 18 Malaysia .218 .464 .124 .115 .032 .047 2.24 16 .154 15 .716 3 Indonesia .222 .312 .236 .129 .018 .084 2.36 13 .160 14 .582 6 Thailand .296 .467 .071 .130 .017 .021 2.09 17 .150 16 .777 1 Philippines .350 .294 .151 .158 .041 .006 2.24 14 .200 13 .648 5 United Kingdom .132 .315 .214 .246 .063 .031 2.79 9 .318 8 .461 11 Ireland .127 .352 .168 .234 .068 .050 2.75 11 .318 9 .505 8 France .187 .281 .178 .193 .136 .025 2.81 8 .337 7 .480 9 Germany .056 .169 .238 .308 .193 .036 3.43 2 .520 2 .233 16 Sweden .088 .257 .172 .192 .256 .036 3.28 3 .464 3 .358 14 Italy .194 .271 .200 .190 .119 .027 2.76 10 .318 10 .477 10 Spain .100 .299 .209 .242 .068 .082 2.87 7 .338 6 .434 13 Portugal .211 .291 .165 .216 .071 .046 2.63 12 .301 11 .526 7 Greece .353 .334 .168 .077 .017 .052 2.02 18 .098 18 .724 2 Mean .173 .295 .195 .216 .081 .040 2.73  .297  .468 Standard deviation .378 .456 .396 .411 .272 .197 1.23  .457  .499 Notes: Trade Opinion gives responses to the following question: “Please tell me how much you agree or disagree with the following statements: [Country] should limit the import of foreign products.”  Average Trade Opinion is the average of Trade Opinion excluding I don’t know (6) and NA (7) answers.  The second column of each variable gives the ranking of countries according to that variable.   88  I am primarily interested in the effects of government credibility associated with the welfare state on trade attitudes.  Before I present a series of empirical models highlighting different types of determinants of trade attitudes in the next section, as a first pass through the data, it is instructive to examine whether there is a systematic association between subjective evaluations of government credibility and levels of protectionism on a cross- national basis.  As argued in the earlier section, government credibility has two different components: (1) trust in government’s ability and (2) trust in its desire to protect people from the vagaries of the free market.  First, I use the survey question that asks how much respondents agree or disagree with the statement that “with regard to most of the big problems we face, what the [respondent’s country] government decides doesn’t make much difference,” as a proxy for trust in government’s ability.  Respondents can answer “strongly disagree,” “disagree,” “neither agree nor disagree,” “agree,” and “strongly agree.”  I scored these responses from 1 to 5, giving those who answered “strongly agree” a 1, and those who answered “strongly disagree” a 5.  Although this question does not ask respondents’ opinions specifically about their government’s ability (or effectiveness) to implement social protection, it still allows us to assess how the respondents evaluate the effectiveness of their government in general – presumably, respondents negatively evaluating their government’s ability to make a difference in one aspect will not make a completely different evaluation of the government’s ability in other aspects.  Second, to measure trust in governments’ desire to provide social protection, I make use of responses to the question asking respondents how proud they are of the welfare system in their country, by assuming that respondents who are proud of their welfare system feel that they are well protected by it, while those not proud feel that they are not.  I assigned a value of 1 to respondents who answered “not proud at all,” and 4 to those who answered “very proud,” and thus, higher values of this variable   89 reflect more positive evaluations of the country’s welfare system.  Presumably, the type of welfare regime people live under – e.g., conservative, liberal or social-democratic welfare state – formulates people’s expectations of their government as to to what extent the government will limit the costs and distribute the benefits of open markets through social welfare policies; and it is highly likely for the government to follow the similar welfare-state trajectory as instructed by the legacy of the country’s previously instituted social policies.  In this regard, a subjective evaluation of the country’s welfare system can serve as a proxy for trust in government’s desire to provide social protection.  In addition, a variable that measures levels of perceived corruption in politics is also used as another proxy for trust in government’s desire.  Presumably, if the publics perceive that corruption is rampant and politicians are primarily motivated by rent-seeking, they would hardly find the government’s redistributive intent trustworthy.  To measure how widespread perceived corruption is among politicians, I make use of responses to question asking respondents how much they agree or disagree with the statement that “there is widespread corruption among those who manage our national politics”.   To be able to make cross-national comparisons, I created the four variables, Goveffect, Welfare, Corruption, and Protectionism, by producing an average of the values of each variable for each country.  Figure 3.1, 3.2, and 3.3 show the results.  Does protectionist sentiment have any systematic relationship to the public perception of government effectiveness on a cross-national basis?  Figure 3.1 shows that the answer is broadly yes.  In Figure 3.1, I plot protectionist sentiments against government effectiveness that I obtained on Goveffect across 18 countries.  Several results stand out.  First, as expected, there is a negative association between protectionist sentiment and the subjective evaluation of government effectiveness   90 on a cross-national basis.  In a simple cross-national OLS regression model with Protectionism as the dependent variable and Goveffect as the independent variable, the coefficient for Goveffect is –.689 with a standard error of .265, which is statistically significant at the 95% confidence level.  This remains almost unaffected even after I control for the impact of international economic integration,16 which is expected to correlate rather closely with levels of protectionism.  With few exceptions, the relationship between Goveffect and Protectionism is negative and linear to a good approximation in each country. This point will be covered in more detail in the next section.  Second, there are other features of the pattern of association reported in Figure 3.1 worth noting.  Interestingly, though perhaps not surprisingly, new (Asian) democracies – e.g., Thailand, Philippines, Indonesia and South Korea – are clustered together: they all exhibit low levels of government effectiveness and high levels of protectionism.  Contrary to the new democracies, countries under authoritarian rule – e.g., China, Singapore, and Malaysia – display relatively high levels of government effectiveness.  The three countries vary in terms of protectionism, but both China and Singapore certainly display relatively low levels of protectionism, compared to the new democracies above.  Third, the two European countries that emerged as the most pro-trade countries – i.e., Germany and Sweden – also comprise of a group displaying the strongest feelings of government effectiveness.       16 To measure the impact of global economic exposure, I used the index provided by Foreign Policy magazine. The index indicates world rankings of eighteen countries in the two respective categories of trade and FDI by percentage of GDP.  The index is available online at www.foreignpolicy.com .   91 Figure 3.1 Scatterplot of protectionism and subjective evaluation of government effectiveness    Notes: The variable, Protectionism, on Y axis indicates average values of protectionism on the scale from 1 to 5.  High values of Protectionism above reflect average support for protectionism of the countries indicated, whereas low values reflect pro-trade attitudes.  The variable, Goveffect, on X axis indicates average values of subjective evaluations of government effectiveness on a scale from 1 to 5.  The points presented above are national averages: for example, Singapore ranked the lowest in Protectionism with a national average of 2.04 and ranked the fourth in Goveffect with an average of 3.03 after Sweden, Portugal, and Germany.    Figure 3.2 plots protectionist sentiments (Protectionism) alongside average subjective evaluations of the welfare system (Welfare).  At first glance, there appears to be no   92 association between the two variables on a cross-national basis.  However, excluding less developed countries17 (LDCs) that have never enjoyed the privilege of extensive social welfare programs, the scattorplot reveals a strong and tight relationship between protectionist sentiment and subjective evaluations of the welfare system on a cross-national basis.  For the 10 developed countries (nine European countries plus Japan) in the data set, the coefficient of Welfare is.589 with a standard error of .272, which is statistically significant at the 90% confidence level.  That is, the more positive the average subjective evaluation of the country’s welfare system, the less protectionist a country, on average.  In general, people who live under a conservative or social democratic welfare regime (e.g., Germany, France and Sweden) tend to be prouder of the country’s welfare system than those under liberal or Mediterranean welfare regime (e.g., UK, Ireland, Italy, Portugal, and Greece).  Consequently, protectionism of social democratic and conservative welfare regimes on average seems to be weaker than that of other types of welfare countries.  When it comes to outliers (LDCs) I suspect that since they have never experienced an extensive welfare (social security) system, a subjective evaluation of the welfare system – i.e., how proud they are of the country’s welfare system – could measure something different, which I will discuss in more detail in the next section.   17 I divide the countries into two groups – (1) a group of developed countries and (2) a group of less developed countries (LDCs).  I include all of the European countries of the data set plus Japan in a group of developed countries, and include the rest in a group of LDCs.   93 Figure 3.2 Scatterplot of protectionism and subjective evaluation of welfare system    Notes: The scatterplot shows a relationship between protectionist sentiment and subjective evaluation of the country’s welfare system.  The variable, Welfare, on X axis reports the average values of respondents’ subjective evaluations of their country’s welfare system on a scale from 1 to 4.  The higher the value, the more proud the respondents are of their welfare system.  The variable Protectionism on Y axis is coded the same as that of Figure 3.1.  The points presented above are national averages: for example, Singapore ranked the lowest in Protectionism with a national average of 2.04 and ranked the second in Welfare with an average of 3.06 after Malaysia which displays the highest average Welfare score (3.17).  Figure 3.3 plots protectionist sentiments (Protectionism) alongside average perceptions of corruption in politics.  As expected, there is a positive association between   94 perceived corruption and protectionist sentiment on a cross-national basis: countries with higher perceptions of corruption tend to display higher protectionist sentiment on average. It is also noteworthy that most of the new democracies are clustered at the top-right corner of the graph: the higher perceived corruption, the higher protectionist sentiment.  Figure 3.3 Scatterplot of protectionism and perceived corruption  Notes: The variable, Protectionism, on Y axis indicates average values of protectionism on the scale from 1 to 5.  High values of Protectionism above reflect average support for protectionism of the countries indicated, whereas low values reflect pro-trade attitudes.  The variable, Corruption, on X axis indicates average values of perceived levels of corruption on a scale from 1 to 5.  The points presented above are national averages.    95 In summary, there are a few cross-national patterns reported in the two figures above worth noting.  First, a subjective evaluation of government effectiveness, as a proxy for public trust in government’s ability, correlates negatively with protectionist sentiment on a cross-national basis.  Second, new (Asian) democracies display relatively low levels of government effectiveness and high levels of protectionist sentiment, whereas countries under authoritarian rule display seemingly the opposite pattern.  Third, pride in the country’s welfare system negatively correlates with protectionist sentiment on a cross-national basis: the prouder people are of their country’s welfare system, the lower the levels of average protectionist sentiment of the country.  The pattern, however, is displayed only among developed countries.  And lastly, a perceived political corruption is positively associated with protectionist sentiment; and new democracies in general display relatively high levels of perceived corruption and high levels of protectionist sentiment.  Then, does government credibility, proxied here by subjective evaluation of government effectiveness and the welfare system, have any systematic effect on protectionist sentiment at the individual level as well?  In the following sections, I will move to the individual level analysis.  3.3.2 Government credibility model  I use an Against-trade Dummy from the ASES data set as a dependent variable and estimate a series of logit models.  The models identify government credibility as crucial to determining an individual’s attitudes toward trade.  As discussed above, government credibility has two components: (1) trust in government ability and (2) trust in government desire (to protect people from the vagaries of the free market).  First, as proxies for trust in government ability, I include the following two variables as well as government effectiveness (Government effectiveness), which I used in the earlier section: a subjective evaluation of government dealing   96 with the economy (Government dealing with the economy); and a subjective evaluation of government dealing with unemployment (Government dealing with unemployment).  If economic insecurity – understood to be an individual’s perception of the risk of economic misfortune that mainly stems from volatile employment situation (Dominitz and Manski 1997; Mughan and Lucy 2002; Scheve and Slaughter 2004) – contributes to the backlash against trade liberalization as it is often argued18, it is expected that people who answered that their government is dealing well with unemployment and the economy are more likely to support trade than those who answered not.  Second, as proxies for trust in government desire, I include the following four variables: (1) a subjective evaluation of the nation’s welfare system (Pride in the nation’s welfare system); (2) a subjective evaluation of the nation’s democracy (Pride in the way nation’s democracy works); (3) a subjective evaluation of the government’s handling of political corruption (Government dealing with corruption); and (4) a perceived levels of corruption in politics (Perceived levels of corruption).  As discussed in the earlier section, presumably regime types which people live under – with respect to both welfare system and form of government – serve as guidance for anticipating to what extent their government is willing to compensate losers from trade19.  Likewise, perceived corruption serves as a proxy to measure levels of government credibility: perceptions of corruption or rent-seeking politicians are basically a reflection of citizens’ lack of trust in government’s desire to implement social policies that aim to protect citizens from the vagaries of open market.  I expect both corruption variables  18 As discussed in the earlier section, the economic openness-welfare spending nexus is based on the argument that increases in economic insecurity from trade generate demands for more generous social welfare spending to compensate losers from trade. 19 Admittedly, however, the indicators suggested above do not have clear-cut boundaries between government ability and desire.  For example, a subjective evaluation of government handling political corruption can also be used as a proxy for trust in government desire as well.   97 – Government dealing with corruption and Perceived levels of corruption – to be positively associated with protectionist sentiment.  The regression results are presented in Table 3.2.  The numbers are the marginal effects of each variable, i.e., the estimated change in the probability of being protectionist given a marginal increase in the value of the relevant independent variable, holding all the other independent variables constant.   98 Table 3.2 Government credibility model Model 1 Model 2 Model 3 Logit estimation DCs (1) LDCs (2) DCs (3) LDCs (4) DCs (5) LDCs (6) Dependent variable Against-Trade Dummy (Protectionism) Trust in government ability     Government effectiveness -0.046 (0.006)*** -0.070 (0.009)*** -0.047 (0.006)*** -0.067 (0.009)***         Government dealing with the economy     -0.017 (0.009). -0.068 (0.012)***     Government dealing with unemployment Trust in government desire     Pride in nation’s welfare system -0.056 (0.007)***  0.043 (0.010)*** -0.065 (0.007)***  0.007 (0.011) -0.052 (0.007)*** -0.006 (0.011)     Pride in the way nation’s democracy works     * Not asked in China      Government dealing with corruption     * Not asked in China  0.015 (0.009). -0.106 (0.011)***  0.007 (0.009) -0.103 (0.010)***         Perceived levels of corruption     * Not asked in China      0.042 (0.006)***  0.073 (0.008)*** Nationalism (Control variable)     National pride   0.117 (0.009)***  0.136 (0.013)***  0.119 (0.008)***  0.148 (0.013)*** Individual attributes     Education -0.064 (0.006)*** -0.008 (0.009) -0.060 (0.006)*** -0.003 (0.009) -0.060 (0.006)*** -0.011 (0.009)     Subjective household income  0.023 (0.009)* -0.043 (0.013)**  0.018 (0.010). -0.044 (0.014)**  0.016 (0.009). -0.049 (0.013)***     Union membership (Dummy) -0.013 (0.014) -0.075 (0.025)** -0.003 (0.015) -0.043 (0.025)  0.005 (0.014) -0.046 (0.025).     Public sector employment (Dummy)  0.032 (0.015)*  0.059 (0.025)*  0.031 (0.016)*  0.045 (0.026)*  0.035 (0.015)*  0.053 (0.025)*     Age  0.001 (0.004)  0.007 (0.008) -0.003 (0.005)  0.010 (0.008) -0.000 (0.005)  0.010 (0.007)     Gender (female)  0.057 (0.013)***  0.058 (0.019)**  0.061 (0.013)***  0.054 (0.019)  0.057 (0.013)***  0.062 (0.018)*** Number of observations  6290  3555  6120  3521  6390  3599 Log Likelihood -4170.793 -2323.066 -3967.975 -2248.031 -4144.258 -2309.898 Notes: The table contains the estimated marginal effect on the probability of being protectionist, given an increase in the value of the relevant regressor, holding all other variables constant.  The standard errors are presented under each marginal effect.  Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1.   Question wording for the control variable is as follows: National pride: “Overall, how proud are you to be [Japanese]?”; The variable is coded as follows: 1 = not proud at all, 2 = not so proud, 3 = somewhat proud, 4= very proud.  Education is coded as follows: 1 = didn’t receive education, 2 = 1-5 years, 3 = 6-10 years, 4 = 11-15 years, 5 = 16-20 years, 6 = 21-25 years, 7 = 26 years or over, 8 = no answer.  Subjective household income refers to the answers to the question of “how would you describe your household’s living standards?” and it is coded as follows: 1 = high, 2 = relatively high, 3 = average, 4 = relatively low, 5 = low, and 6 = no answer.  Union membership equals 1 if the respondent is a member of trade union, 0 if he is not.  Public sector employment equals 1 if the respondent is employed in the state sector, 0 if he is not.  Age is coded as follows: 1 = Age10 , 2 = 20s, ,..., and 7 = 70s.    99 Table 3.2 Government credibility model (Continued) Model 4 Model 5 Logit estimation DCs (7) LDCs (8) DCs (9) LDCs (10) Dependent variable Against-Trade Dummy (Protectionism) Trust in government ability     Government effectiveness     Government dealing with the economy     Government dealing with unemployment -0.014 (0.008). -0.073 (0.011)*** -0.013 (0.008) -0.073 (0.011)*** Trust in government desire     Pride in nation’s welfare system   -0.054 (0.007)*** -0.005 (0.011)     Pride in the way nation’s democracy works     * Not asked in China -0.043 (0.008)*** -0.002 (0.011)      Government dealing with corruption     * Not asked in China      Perceived levels of corruption     * Not asked in China  0.047 (0.006)***  0.072 (0.008)***  0.045 (0.006)***  0.073 (0.008)*** Nationalism (Control variable)     National pride  0.125 (0.009)***  0.136 (0.013)***  0.116 (0.008)***  0.141 (0.013)*** Individual attributes     Education -0.062 (0.006)*** -0.009 (0.009) -0.060 (0.006)*** -0.010 (0.009)     Subjective household income  0.016 (0.010). -0.042 (0.014)**  0.017 (0.009). -0.046 (0.013)***     Union membership (Dummy)  0.012 (0.015) -0.057 (0.026)*  0.010 (0.014) -0.044 (0.025).     Public sector employment (Dummy)  0.026 (0.015). 0.052 (0.026)*  0.031 (0.015)  0.054 (0.026)*     Age -0.001 (0.005)  0.012 (0.008) -0.000 (0.004)  0.011 (0.008)     Gender (female)  0.064 (0.013)***  0.055 (0.019)**  0.055 (0.013)***  0.062 (0.019)*** Number of observations  6346  3462  6624  3589 Log Likelihood -4126.099 -2224.588 -4330.739 -2298.648    Regressions (1), (3), (5), (7), and (9) include only developed countries; and regressions (2), (4), (6), (8), and (10) include only LDCs.  Trade theory tells us that trade liberalization affects the welfare of workers and the poor differently between developed and less developed countries.  The Heckscher-Ohlin theory, for example, suggests that the owners of relatively abundant factors of production benefit from trade; and thus we expect to find different determinants of trade opinions between LDCs characterized by lack of   100 capital and abundance in low-skilled labour, on one hand, and developed countries characterized by abundance in highly skilled labour and capital, on the other hand.  I include individual attributes in every model, and the empirical patterns found correspond to the Heckscher-Ohlin theory: individual skills proxied by levels of education correlate negatively with protectionist sentiment in developed countries, but the correlation is weak and even reversed in LDCs, and union membership negatively associates with protectionist sentiment in LDCs while there is no such association in developed countries.  Given the different impacts of trade liberalization on the national economy, as well as the contextual differences in terms of welfare state legacy and system of government (democracy vs. authoritarian) between developed and less developed countries, I suspect that the size (or even direction) of the effects that the key variables have on protectionist sentiment also differ.  I find evidence of a strong effect of government effectiveness on protectionist sentiment, which survives all specification I have tried: both in developed and less developed countries, people are less likely to support protectionism as they evaluate their government’s effectiveness more positively.  When the variable Government effectiveness is replaced with Government dealing with economy or Government dealing with unemployment, I find little change in the outcome: respondents with positive evaluations of government dealing with the economy and unemployment are more likely to support trade.  The results basically serve as strong evidence that public trust in their government doing a good job on the economy or unemployment or public perception of government effectiveness in general plays a significant role in garnering support for trade.    101 I also find evidence of a strong effect of pride in the nation’s welfare system on protectionist sentiment; but interestingly, the association is negative in developed countries, while it is positive in LDCs.  Both coefficients are statistically significant at the 99% confidence level.  That is, those who are proud of their welfare system are less likely to be protectionist in developed countries, but more likely to be protectionist in LDCs.  In DCs, pride in the country’s democracy as well positively correlates with protectionist sentiment. But as shown in Models 2, 3, 4, and 5, the association between pride in nation’s welfare system (and democracy) and protectionist sentiment in LDCs disappears once nationalism (proxied by National pride, pride in the nation’s political influence in the world, and pride in the nation’s armed force) are included, whereas it survives in developed countries.  Given that most of the LDCs have never experienced expansive social welfare protection, asking respondents how proud they are of the country’s welfare system may not have measured respondents’ trust in government’s ability; rather it may have simply measured how proud they are of their country in general.  As expected, the two corruption variables – Government dealing with corruption and Perceived levels of corruption – positively associate with protectionist sentiment.  More specifically, 1 point increase in corruption perceptions on a scale from 1 (lowest) to 5 (highest) increases protectionist sentiment by about 4 percent in developed countries (DCs) and about 7 percent in LDCs.  Likewise, 1 point positive increase in evaluations of government dealing with corruption decreases protectionist sentiment by about 10 percent in LDCs, while no such effect is found in DCs.  In general, the impact of corruption perceptions on protectionist sentiment is found stronger in LDCs than DCs.    102  Overall, the micro results confirm the significance of government credibility. Confidence in government effectiveness and economic performances negatively associates with protectionist sentiment at the individual level; and the negative effect is strong both in developed and less developed countries.  Pride in the nation’s welfare system also negatively associates with protectionist sentiment at least in developed countries.  From these individual-level results put together with the cross-national patterns reported in the earlier section, one can infer that the embedded liberalism compromise may not be made in LDCs. As seen in Figure 3.1, new democracies are clustered together, displaying low levels of government effectiveness alongside high levels of protectionist sentiment. I also find government credibility and protectionism are negatively correlated at the individual level.  As discussed earlier, the embedded liberalism compromise is based on the premise that people trust their government’s desire and ability to cushion trade-induced insecurity.  If the public has little trust in government, which is actually the case in most of the LDCs, the public would not be willing to compromise on trade liberalization policies in return for the government’s welfare promise.  Also, unless welfare spending garners public support for trade, there is not much incentive for the government to increase welfare spending.   In fact, the core assumption of the embedded liberalism thesis is that trade liberalization increases economic insecurity; and economic insecurity in turn develops policy attitudes against trade.  It is argued that governments committed to trade liberalization invest in social welfare, hoping to maintain or build public support for trade liberalization policies by virtue of the security-enhancing effects of social welfare.  I suspect, however, that   103 economic insecurity defined narrowly as fear of personal job loss,20 may not be the main determinant of trade attitudes.  It may just be that people are moved by their perception that their country as a whole will have a harder time finding jobs – that is, differing degrees of sociotropic economic insecurity may be a better predictor of trade attitudes.  It may also be that uncertainty about economic performance, as well as economic insecurity, shapes policy attitudes against trade.  Or even unspecified worries about the country may magnify protectionist sentiment.  If that is the case, this will strengthen the significance of government credibility.  If it is the case that personal (egocentric) economic insecurity is the key to shaping policy attitudes against trade, offering compensation to losers from trade at time t could be able to garner support for trade at time t +1.  If sociotropic insecurity also matters, however, welfare compensation at time t may have no discernable impact on trade attitudes at time t +1, because such insecurity cannot be easily mitigated in a short time frame by material compensation directed to the losers from trade.  What will shape trade attitudes now is the level of government credibility, which can only be built over a long period of time because credibility is a matter of perception.  It does not matter much whether it is trade liberalization that causes such insecurity; what matters is that unspecified worries about the country itself, the national economy, or even national politics shape policy attitudes against trade.   Table 3.3 confirms the significance of sociotropic insecurity.  Sociotropic insecurity, including worrying about your country, the country’s economy and unemployment, and the country’s corruption, is a highly significant determinant of protectionist sentiment.  20 Mughan and Lacy (2002) argue that it is the norm in studies of economic insecurity to define insecurity simply in terms of fear of job loss, by citing Witte (1999) and Dominitz and Manski (1997) (De Witte 1999, 155; Dominitz and Manski 1997, 261-287; Mughan and Lacy 2002, 513-533).   104 Sociotropic insecurity matters more than egocentric insecurity: egocentric insecurity has only minor impact on protectionist sentiment in developed countries, and almost no impact on protectionist sentiment in LDCs after controlling for sociotropic insecurity (Model 2).  Table 3.3 also shows that after controlling for the variables proxied for trust in government, the marginal effects of insecurities on protectionist sentiment go down (Model 3 and 4).  For example, when I include the trust variables – government effectiveness, and perceived levels of corruption – the marginal effect of sociotropic insecurity (worry about your country) decreases by 1.5% from 7.2% to 5.7% in LDCs (Model 3).   Model 5 is a logit model including interactions between government effectiveness and sociotropic insecurity and between sociotropic insecurity and LDCs.  As it is not straightforward to discern how the variables – sociotropic insecurity, government effectiveness, and LDCs – combine to influence the probability of being protectionist, I made use of I made use of effect displays in R by John Fox’s effects package in R to visualize the interaction effects in Model 5 (Fox 2003).  Figure 3.4 shows the outcomes.   Table 3.3 Models of insecurity Model 1 Model 2 Model 3 Model 4 Model 5 Logit estimation DCs LDCs DCs LDCs DCs LDCs DCs LDCs All data Dependent variable Against-Trade Dummy (Protectionism) Egocentric Insecurity    Worry about your work situation  0.029 (0.007)***  0.028 (0.007)***  0.021 (0.007)**  0.010 (0.008)  0.024 (0.008)**  0.018 (0.009)*  0.020 (0.008)*  0.008 (0.009)  0.022 (0.006)*** Sociotropic Insecurity    Worry about your country  0.067 (0.008)***  0.072 (0.008)***    0.051 (0.008)***  0.057 (0.010)***    0.014 (0.017)    Worry about the country’s economy    0.075 (0.009)***  0.066 (0.010)***    0.065 (0.009)***  0.044 (0.012)***     Worry about the country’s unemployment    0.026 (0.009)**  0.068 (0.010)***    0.020 (0.009)*  0.067 (0.011)***  Trust in government     Government effectiveness  -0.041 (0.005)*** -0.050 (0.007)*** -0.039 (0.005)*** -0.042 (0.007)*** -0.078 (0.011)***     Pride in welfare  -0.031 (0.006)*** -      Perceived levels of corruption     * Not asked in China   0.030 (0.005)***  0.052 (0.006)***  0.031 (0.005)***  0.047 (0.005)***  0.041 (0.004)*** Less developing countries    LDCs (Dummy: 1 = LDCs, 0 = DCs)     0.071 (0.025)** Interaction term     Worry about your country:     Government effectiveness       0.015 (0.005)**    Worry about your country:    LDCs       0.009 (0.012) Number of observations  9217  7437  9087  7430  8217  6112 8156  6210  14542 -2 Log Likelihood -6261.432 -5053.636 -6143.96 -5002.187 -5499.666 -4030.677 -5442.432 -4010.263 -9688.514 Notes: The table contains the estimated marginal effect on the probability of being protectionist, given an increase in the value of the relevant regressor, holding all other variables constant.  The standard errors are presented under each marginal effect.  Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1.  Question wording for each variable is as follows: Worry about your work situation: “How worried are you about your work situation?”; Worry about your country: “How worried are you about your country?”; Worry about the country’s economy: “How worried about the economy?”; and Worry about unemployment: “How worried about unemployment?”  The variables are coded as follows: 1 = not worried at all, 2 = somewhat worried, 3 = very worried.  In Model 3, I did not include Pride in welfare for LDCs, because the analysis in the earlier section confirmed that Pride in welfare serves as a proxy for nationalism (rather than evaluations of welfare system) in LDCs.   106 Figure 3.4 Interaction effects between sociotropic insecurity and government effectiveness  Notes: Effect display for the interaction of SocioInsecurity and GovtEffectiveness in the logit model fit to the total ASES data.  The vertical axis is labelled on the probability scale, and a 95 percent pointwise confidence interval is drawn around the estimated effect.  LDCs is a dummy variable (LDCs = 1 and DCs = 0); and the marks in orange color show the levels of GovtEffectiveness (lowest (1), middle (3), and highest (5)) on a scale from 1 to 5.   Figure 3.4 displays graphs of the government effectiveness by sociotropic insecurity by LDCs effect, as fitted in Model 5 (Table 3.2).  The three graphs on the top show the cases of LDCs, while the bottom three graphs show the cases of developed countries (DCs).  The   107 two graphs on the far left display how the probability of Protectionist sentiment changes alongside SocioInsecurity when Govteffectiveness is held constant to the lowest level; the two graphs in the middle and the two on the far right display the changes when Govteffectiveness is held constant to the middle level or to the highest level, respectively.  Several of results stand out.  First, across the board, respondents in LDCs are a lot more likely to be protectionist than their counterparts in DCs.  All three lines on the top three graphs are located above their counterparts on the bottom three graphs.  Second, Govteffectiveness has a discernable mitigating effect on the probability of being protectionist: as Govteffectiveness increases, the probability of being protectionist goes down.  Third, the mitigating effect of Govteffectiveness however seems weaker as SocioInsecurity increases.  The fact that the slop of the lines becomes steeper as Govteffectiveness increases suggests that while Govteffectiveness mitigates the probability of being protectionist at all levels of SocioInsecurity, its impact is stronger among those who display a relatively mild level of SocioInsecurity (than among those who display a high level of SocioInsecurity).  In short, the results displayed in Table 3.2 and Figure 3.4 can be summarized as the following two: (1) sociotropic insecurity plays a more significant role in explaining protectionist sentiment than egocentric economic insecurity; (2) government effectiveness effectively mitigates protectionist sentiment, although its importance as a predictor decreases for those whose sociotropic insecurity levels are higher.  The importance of sociotropic insecurity relative to egocentric economic insecurity in explaining protectionist sentiment strengthens the argument that perceptions, more specifically, perceptions of government commitment to cushion miscellaneous insecurities that people already have or that people think trade might breed, are the key to garnering public support for trade.  This implies that   108 only when people view their government as credible, so much that they feel they are well protected by the government, would they be willing to compromise on further trade liberalization.  Government welfare spending would have little discernable impact on trade attitudes in countries that lack such credibility.   3.4 Conclusion  The embedded liberalism thesis involves a grand bargain whereby the public agrees to open markets and a government in exchange promises to moderate the volatility of open markets and provide social safety nets as compensation.  For this grand bargain to be enacted and sustained, the public should be willing to compromise on trade liberalization policies in return for the government’s welfare promise.  I paid attention to a problem of credible commitment that is likely to arise here, because trade liberalization and a government’s compensation often involve intertemporal, non-simultaneous exchanges between the government and the public.  Given the temporal feature of the grand bargain, people are likely to compromise only when the government’s promises to cushion the adverse domestic effects of open markets are viewed as credible.  If viewed as not credible, the grand bargain would not be enacted, let alone sustained. In LDCs that have never enjoyed the privilege of expansive social protection and that are often characterized by lack of effective government institutions, security-enhancing functions of government promises are not likely to work at the same level as they do in countries with an effective government and a long welfare state tradition.    109  The empirical tests confirm the significance of government credibility.  I find evidence of a strong effect of government credibility on protectionist sentiment both at the cross-national and individual level of analysis.  I also find that not egocentric economic insecurity but sociotropic insecurity is the key to policy attitudes against trade.  The significance of sociotropic insecurity as a micro-foundation of the embedded liberalism thesis strengthens the argument that perception matters: it is highly likely that material compensation itself that is directed toward losers from trade has no discernable security- enhancing functions in countries where there is little trust in government.  Government credibility matters, but it takes long time to develop.  The relationship between trade and welfare state, thus, can only be long-term and historically contingent.      110 4 TRUST IN GOVERNMENT AND PROTECTIONIST SENTIMENT: KOREAN PUBLIC OPINION ON THE KOREA-US FREE TRADE AGREEMENT (KORUS-FTA)  4.1 Introduction: Framing the puzzle  Public opinion polls on the subject of trade often reveal two interesting dichotomies.  The first is a dichotomy between economists and the general public: while few economic policies command as much consensus among economists as the benefits of trade liberalization, the public in most industrialized countries has been largely sceptical (Caplan 2002, 433-458; Fuller and Geide-Stevenson 2003, 369-387; Irwin 2005b).  The public is not inherently and consistently sceptical of trade, however, and this creates a second dichotomy: between the views of trade held by the public when the issue is viewed in the abstract versus when there is a particular trade policy to consider.  While the public displays generally positive attitudes toward trade in principle, they are largely ambivalent about specific trade policy initiatives (Irwin 2005b; Scheve and Slaughter 2001; Stokes, Choate, and Council on Foreign Relations 2001).  The quadrennial surveys of the Chicago Council on Global Affairs (CCGA) have consistently shown that majorities around the world have a generally positive view of globalization and believe that international trade has a positive impact on national economies (2002, 2004, and 2007).  When the public is asked about specific trade policy initiatives, however, public support is considerably lower.  These general findings have been fairly consistent across countries.   Korea is one of the countries where the dichotomy between a willingness to accept increased international trade, and a hesitation to support integration driven by specific policy   111 initiatives, is most pronounced.  As an export-oriented economy where trade has played a major role in boosting the country’s rapid economic growth over the last few decades, Korea is one of those countries that display the highest levels of public support for trade liberalization and a strong consensus on the positive impact of trade on national economies. Indeed, among the 17 countries surveyed by the CCGA in 2007, Korea ranked the second highest next only to China in its public support for trade liberalization, recording 86 percent21.  When it comes to the government’s specific trade policy initiatives, however, the support of the Korean publics seems lukewarm at best.  As of 2011, Korea has signed the two major free trade agreements – one with the United States and the other with the European Union – which remain pending for ratification.  Although the public support for each FTA has fluctuated during the negotiation process, it is in general not comparable to the public’s support for trade in principle, which is overwhelmingly positive.  According to the public opinion polls that were conducted right after the European Union - Korea Free Trade Agreement (EU-KOREA FTA) was signed in 2009, only about half of the respondents (49.4%) agreed that the deal should be ratified in the National Assembly.  The KORUS FTA as well gained support from only half of the respondents (50.1%).22  At the same time, it is also one of the countries where protest against trade liberalization has been most vigorous, as one can see from the country’s brouhaha over US beef imports in 200823.  21 The survey question used to measure support for globalization is as follows: “Do you believe that globalization, especially the increasing connections for our economy with others around the world, is mostly good or mostly bad for [country]” (2007, CCGA).  For the full report of the survey, find http://www.worldpublicopinion.org/pipa/pdf/apr07/CCGA+_GlobTrade_article.pdf. 22 The survey was conducted by Seoul-based research company Realmeter in July, 2009.  The respondents were asked if they would support for the ratification of each of the free trade deals – the EU-Korea FTA and the KORUS FTA.  For the EU-Korea FTA, 49.4 percent of the respondents supported its ratification, whereas 24.8 percent opposed it, and the rest refused to answer or answered they did not know.  The respondents’ answers to KORUS FTA revealed almost the same pattern: 50.1 percent supported it, and the rest were either ambivalent about or opposed to it with 19.2 percent and 30.7 percent, respectively. 23 President Lee Myung Bak's removal of restrictions on U.S. beef has plunged his administration into a crisis that could imperil a free-trade pact with the U.S.  (Businessweek June 9, 2008)   112   This chapter aims to identify the sources of this divergence in Korea, support for trade in the abstract and scepticism about specific trade policies, bilateral free trade agreements in particular.  My basic argument is that a lack of political trust, one of the main characteristics of new democracies, is an important source of the divergence.  Governments’ specific trade policy initiatives are distinctive from the abstract principle of trade liberalization in the sense that “politics” is inevitably involved in the process of implementing the former.  Trade policy initiatives, trade negotiations in particular, are a political process: the government representatives need to coordinate the interests of domestic constituencies; and they also have to deal with bargaining in a context of international negotiations that are often characterized by an asymmetry of power (Putnam 1988a, 427; Cameron and Tomlin 2002).  Not only that, differences in domestic political institutions – e.g., whether it is centralized or divided – and availability of non-agreement alternatives all likely play a role in trade negotiations (Cameron and Tomlin 2002).  I argue where politics is involved is where political trust becomes important.  In democracies, most of which are representative in nature, the government’s functioning as a reactor and coordinator of diverse domestic interests or an international negotiator depends heavily on public trust.  If public trust in political institutions (or individual representatives) declines, legitimacy of a given administration, or even the regime itself, will be called into a question. Somewhat ironically, a transition to democracy and the resulting increase in civic engagement have decreased political trust in new democracies.  Korea is not an exception to this trend, which I argue is primarily responsible for the lukewarm support for the specific trade initiatives despite the overwhelmingly positive attitudes toward trade in principle.    113 Political trust, defined here as public sentiments toward the responsiveness of the political process and effective governance, is a reflection of how citizens perceive government institutions and individual representatives relative to their expectations (Hetherington 2005). Given the importance of perceptions, it is not surprising that new democracies suffer from a declining level of political trust.  On the one hand, citizens’ expectations about government are likely to increase with democratization; but on the other hand, increased civic engagement in a newly democratized country likely contributes to diminishing political trust by exposing the public to the illegitimate and corrupt practices of government institutions on a more regular basis (Espinal, Hartlyn, and Kelly 2006, 200-223).  Indeed, Catterberg and Moreno (2006), in their analysis of the trends in political trust in new and old democracies over the last 20 years, find that a considerable number of new democracies have shown a lower level of political trust in recent years than in the past.  This is very noticeable among Koreans (Catterberg and Moreno 2006, 31-48).  Park (2004) likewise demonstrates that the percentage of Koreans who trust government institutions has dropped 30 percent over the past decade (Park 2004).  Scholars of democratic consolidation have pointed out that political pact-making processes between authoritarian elites and political opposition, which often served as a successful formula for democratic transition, could limit democratic consolidation of the country in the long term (Hagopian 1990, 147; Munck and Leff 1997, 343-362).  Indeed, the negative legacies of Korea’s path of democratic transition, i.e., “democracy by undemocratic means” if borrowing Hagopian’s words, have impeded the institutionalization of democratic rules and procedures (Lee 2007, 99-125).  Given this nature of Korean democracy, a prototype of what O’Donnell named “a delegative democracy,” i.e., a mixture of selected democratic norms of majoritarian rule and   114 authoritarian practice it is not surprising to observe the country display low levels of political trust (O’Donnell 1994).   Little research has been conducted on how political trust affects the formation of public policy preferences, however.  Admittedly, not all types of policy issues require political trust (Hetherington 2005).  While contrasting public policies that everyone benefits from, e.g., environmental spending or national defence, with those that require some segment of society to make sacrifices without promising any discernable, immediate benefits, Hetherington in his pioneering work on political relevance of political trust argues that a lack of political trust ought to affect public support for the latter but not for the former (Hetherington 2005). Hetherington further demonstrates that a lack of political trust accounts for the puzzling discrepancy between (1) increasingly conservative American redistributive programs and (2) American public opinion of these programs that is increasingly liberal, i.e., more supportive of redistributive spending, over the last 20 years.  In other words, he argues that it is not public conservatism, but a lack of political trust that led to conservative redistributive programs in the United States.  Also, political theorists have connected risk and political trust (Warren 1999).  Warren (1999) argues that new policy initiatives which entail a great deal of uncertainty and risk are less likely to be implemented without public trust in the government mitigating the risk.  Offe (1999) likewise argues that for high-risk policies where there is a tension between opportunity and risk, trust is indispensible to easing the tension.  There are a good number of reasons why political trust affects specific trade policy preferences as well as redistributive social programs discussed by Hetherington.  First of all, trade policies are distributive in nature.  They generate losers and winners.  As discussed in   115 the earlier chapter and evidenced by findings of behavioural economists, a sense of fairness commonly underpins motivation of individual policy support.  If policies require a segment of society to make sacrifice, those policies increase the need for trust, i.e., trust in the government’s willingness and capacity to craft policies that serve the general welfare.  Indeed, scholar have pointed out that a legacy of patronage-based government hinders new democracies from developing broad policy credibility, by generating suspicion that elected politicians might provide patronage (Keefer 2007, 804-821; Keefer and Vlaicu 2008, 371).  In short, given the distributive nature of trade policy, trust is important for fairness-motivated individuals to ease suspicion of the government providing patronage.  Second, trade liberalization policies entail adjustment costs, which often take the form of increased aggregate unemployment.  Accordingly, as evidenced by the political economy literature, trade liberalization often increases economic insecurity (Hays, Ehrlich, and Peinhardt 2005, 473; Anderson and Pontusson 2007, 211-235; Mughan and Lacy 2002, 513-533).  As discussed above, trust in government is particularly important for garnering public support for the policy where there is risk involved, and trade liberalization is one such policy.  Lastly, trust is also consequential for trade policies because trade policies by nature tend to reinforce the idea to the public that the government works on their behalf.  Scholars have pointed out that most democracies are representative in nature, which is basically why trust matters (Irwin 2005b; Hetherington 2005; Gamson 1968; Bianco 1994).  The presence of a third party in the policymaking process – as is often the case with bilateral or multilateral trade agreements – likely reinforces the representative relationship between the public and the government, or the public’s perception of the government; and in this regard, a low level of trust may limit their representatives’ leeway in making policy decisions on the nation’s behalf.    116  This chapter demonstrates that it is a lack of political trust that generates growing scepticism of specific trade policy initiatives in Korea despite the overwhelming support for the idea of trade liberalization in principle.  To test the relationship between political trust and individual preferences over specific trade policy initiatives, I pursue discourse analysis. This chapter basically examines how the progressive news media, often perceived as strong opponents of free trade, set the agenda on free trade during trade negotiations.  I hypothesize that in new democracies, trade protectionist arguments during trade negotiations are framed along storylines that increase public suspicions of the government’s responsiveness and effectiveness.  Under the premise that the media mirror public opinion, I test the hypothesis vis-à-vis three plausible alternative hypotheses – (1) identity bias, (2) nationalism, and (3) anti-globalization – which will be discussed in detail in the following sections.  More specifically, I examine how Korean progressive newspapers, which are known to advocate anti-free trade positions, developed the agenda on KORUS FTA (Korea- US free Trade Agreement) during the negotiation period from January 2006 to April 2007 and on re-importation of U.S. beef from April to June in 2008.  This chapter is organized as follows.  In the next section, I define political trust and discuss why support for trade policies requires political trust.  Here I review the scholarly discussion on its definitions and its effects on the formation of public policy preferences.  In the following section, I track the pattern of erosion of political trust in new democracies, with particular attention to the case of Korea.  In the later sections, the data are presented and the hypothesis is tested.  I then conclude by discussing the implications of the results.    117 4.2 Political trust and protectionist sentiment  Political trust is a complex concept.  The definition of political trust has numerous variants: it can be diffuse or specific, and instrumental or normative (for detailed review, see Blind 2006).  Easton (1965) defines trust as an evaluative orientation directed toward a political system.  He divides a political system into the “regime” and the “authorities,” and distinguishes trust toward the former from trust toward the latter (Easton 1965).  Diffuse trust refers to the public’s evaluation of the “regime,” the institutional structures of government, while specific trust is directed toward the “political authorities,” the elected officials of a government.  This distinction is important because distrust toward the authorities will be resolved through electoral replacement, but distrust toward the regime likely persists regardless of who is elected, thus calling legitimacy into question (Miller 1974, 951-972; Citrin 1974, 973-988; Hetherington 1998, 791-808; Keele 2005, 873-886).  While diffuse and specific trust is a categorization based on the objects towards which trust is directed, political trust can also be categorized in terms of its motivations – whether they are rational/instrumental or normative.  Warren (1999) for example conceptualizes it as citizens’ willingness to “accept vulnerability to the potential ill will of others by granting them discretionary power over some good” (Warren 1999).  This definition is instrumental, as it assumes convergent interests between citizens and representatives – i.e., by this definition, political representatives are considered trustworthy to the extent that they attend to citizens’ interests (Warren 1999).  Classical political theorists, on the other hand, take a normative view of political trust, and define it as an assessment of the moral values associated with political institutions and authorities (Fukuyama 1995, 89-103; Mara 2001, 820-845).  In political trust based on normative concerns, citizens expect ethical qualities of their representatives and political institutions.   118   For the purposes of this chapter, I conceptualize political trust as an orientation toward political institutions in general (diffuse trust), rather than as individual political incumbents (specific trust), while incorporating both instrumental and normative aspects into the concept.  Scholars have demonstrated that in American politics, specific trust also matters by showing that trust changes in accordance with partisan control of the government (Keele 2005, 873-886). However, it should be noted that political trust is necessarily broader in a conceptual sense.  Political trust is something that transcends partisanship or ideology, exerting an independent influence on the formation of policy preferences (Hetherington 2005).  Given that political trust is stable and consistent at the individual level as comparable to one’s partisanship and ideology (Jon A. Krosnick 1991, 547-576; Hetherington and Globetti 2002b, 253-275), it should be more important than an evaluation of government performance at a given point in time.  In addition, as an independent predictor of support for public policies, diffuse trust necessarily includes both normative and rational expectations of the public.  Clearly, support for government policies likely depends on the extent to which the public trusts the government to do the right thing, i.e., something that people want it to do and/or they perceive as morally/ethically correct, and to do it well.  More specifically, taking the above-mentioned variants of the definition of trust into consideration, I define political trust as public evaluations toward the responsiveness of the political process and effective governance.  Do people think that the policymaking process is responsive and transparent enough that they can make their voice heard?  Do people think that government institutions are capable, effective, and efficient enough to carry out what they ought to do?  This chapter focuses on these traits of political institutions in order to measure levels of political trust.    119  One point that is imperative to understand in this definition is the importance of perceptions (Hetherington 2005).  The public’s evaluations of government are not a direct reflection of the government’s actual performance.  If the public’s expectations are high, political trust will be lower than it should have been otherwise; and/or a negative news media coverage could also result in public misperceptions of government.  What matters is that the perceptions are more important than the reality for garnering policy support from the public.  This implies that levels of political trust depend on the extent to which the government succeeds in building the impression of being trustworthy among the public, regardless of whether politicians’ position-taking in policymaking processes is indeed “democratic responsiveness” motivated by a principled commitment to a delegate model of representation, or whether it is just strategic “pandering” or “simulated responsiveness” (Lippmann 1955; Page and Shapiro 1992; Jacobs and Shapiro 2000).  Likewise, it is important for the government to look competent; indeed, it is probably more important for a government to appear more competent than it actually is.   While there is little doubt that political distrust influences the public’s policy support by breeding conditions in which the government cannot govern effectively (Gamson 1968), little empirical investigation has yet been made into the consequences of political trust/distrust on support for specific government policies.  The only exception is Hetherington and Globetti’s (2002) research on racial policy.  They argue that trust ought not be consequential for all policy issues, but is consequential for policies which “offer no discernible, immediate benefit and which may demand sacrifices,” such as racial policy (Hetherington and Globetti 2002b, 253-275).  Indeed, Hetherington demonstrates that political trust is associated with support for redistributive spending such as antipoverty and   120 race-targeted programs, but not with support for spending on social security, defense, and crime prevention (Hart and Shaw 2001).   Then what about trade policies?  Trade policies indeed fit the types of policies Hetherington categorizes as those in which trust matters: they are distributive in nature, and they generate losers and winners.  On top of that, trade policies involve a great deal of uncertainty (regarding the distribution of gains and losses from trade liberalization). Opening markets to free trade leaves people vulnerable to the vagaries of the international market, and thus puts them more at risk of loss and more uncertain about the policy outcomes (Fernandez and Rodrik 1991, 1146-1155).  In his edited volume, Democracy and Trust, Warren (1999) indeed connects risk and trust by pointing out that when new policy initiatives are successful in focusing on the uncertainties of their outcomes, the public is less likely to extend the trust necessary for the policy implementation (Warren 1999).  This implies that new policy initiatives that inevitably involve uncertainties and risks such as trade policies demand more trust from the public than those whose outcomes are relatively certain. Without such trust, fears of uncertainty are likely to be magnified, thus making the public turn away from the policy initiatives.  In the same volume, Offe (1999) also argues that for high-risk policies where there is a tension between opportunity and risk, the tension must be eased by trust (Offe 1999).  Trust matters for trade policies because they are by nature not domestic policies – i.e., a third party is involved in the policy-making process.  Bilateral or multilateral trade agreements are indeed conceived as a two-level game, consisting of simultaneous negotiations at both the national (domestic) level and the international level (Putnam 1988b,   121 427).  The presence of a third party in the policymaking process often evokes a sense of “common fate”24 among the public – i.e., a sense that one’s fellow citizens are equally affected by the outcomes of trade bargaining (realistic group theory) and have similar concerns about the country’s standing and status relative to its trade partners (social identity theory) – thereby intensifying the perceived representative relationship with their government.  This explains why trade policies require trust.  Trust is essential to the representative relationship: most democracies are representative in nature; it is trust that gives representatives the leeway to make decisions on the nation’s behalf (Bianco, William T. 1994; Hetherington, Marc J. 2005).  In other words, representation demands that citizens trust that their government institutions are working in the country’s best interest. Without such trust, the public would not support the policies, and instead would seek for more direct control over policy decisions.  In contrast, to governments, the logic of a two-level game implies that public commitments could reduce the set of possible bargaining outcomes, thus making bargaining failure more likely.  That is, governments may perceive that greater the public commitments, the higher the audience costs.  Governments committed to free trade would therefore prefer to maintain flexibility in the ongoing negotiations by minimizing the costs incurred by public involvement.  In this regard, trade policies require political trust more than any other public policies.  When political trust is absent, citizens will want to have  24 There are reasons to think that trade negotiations would be framed along the storylines that enhance such group identity.  The mass media today exerts an interesting influence on people’s perceptions of mass collectives.  According to Mutz, the mass media facilitates the influence of anonymous others by creating portraits of the opinions of large collectives.  Research on the effects of the mass media has in fact suggested that its primary impact is on perceptions of policies (or problems) at the collective level, and that people are responding to a “media-constructed pseudo-environment” rather than their immediate personal experiences or self-interests” (Mutz 1998).  Also, especially when the public is exposed to information about ongoing trade negotiations, whether or not their country makes a good bargain with its trade partner is likely to be of concern to them.  Trade negotiations are in fact often framed along storylines that enhance individual self-perception as an in-group member (the nation) interacting individually or collectively with the out-group (its trade partner); so it is no surprise that collective identity is transcendent in this context.    122 direct control over policy decisions, and in contrast, governments will try to hide detailed information from their citizens so as to minimize audience costs, all of which is likely to create a vicious circle promoting both political distrust and protectionism.   In fact, with a few exceptions, most of the existing research on trade policy opinions stays in the domain of (political) economists.  By assuming that individuals judge policies based on their beliefs about whether they personally might gain or lose from the policies, economists have focused their efforts almost exclusively on identifying who in society wins and loses when the policies are implemented (Mayda and Rodrick 2005, 1393; Baker 2005, 924; Scheve and Slaughter 2001).  Theoretical and empirical models of public opinions about trade policies draw materially from economic trade theories such as the Heckscher-Olin (H- O) and Ricardo-Viner (R-V) models – i.e., trade policy preferences are determined by either the factors individuals hold or the industry in which they work.  Political economists’ efforts to link the material consequences of trade to individual preferences over trade policies is, however, incomplete in the sense that they are largely apolitical and centred on material- interest.  Individuals do not necessarily have enough information to figure out the likely distributive consequences of trade policies; given costs attached to the procurement and analysis of political information, becoming informed about the details is not even considered rational (Downs 1957).  Moreover, gains and losses from trade policies are expected but they are never certain.  This all provides room for politics to play a role in the process of opinion formation.  It is indeed argued that while feelings of uncertainty and risks about policies such as trade policy promote attention from the public – i.e., making people more likely to reconsider their previously learned routines, to collect more information, and to change their attitudes toward the policies (Marcus, Neuman, and MacKuen 2000) – they also make people   123 more suspicious of the attributes of politicians and of the system as a whole25.  The absence of political trust makes politicians more vulnerable to such suspicions because distrust acts as an anchor for suspicion (McGraw, Milton Lodge, and Jones 2002, 362-383).  In short, trust is consequential to trade opinion formation: those who are less trusting should be more prone to suspicion, thereby becoming more susceptible to negative messages of the policy in question; and trade policy is one of those policies that subjects individuals to risky or uncertain outcomes, thereby making trust essential.  4.3 Decline in political trust in new democracies  Trust in political institutions has been considered as vital to the consolidation of democracy. Paradoxically, however, distrust prevails and has even been reinforced in many new democracies, which is allegedly the predicted legacy of authoritarian rule and its consequent democratic transition.  Larry Diamond (1999) in Developing Democracy toward Consolidation suggests three generic tasks that new democracies must handle if they are to become consolidated: (1) democratic deepening, (2) political institutionalization, and (3) regime performance.  Ensuring the latter two – political institutionalization and regime performance, which appear to be interconnected – is particularly important for building and reinforcing political trust.  Such tasks, however, are challenging for most of the new democracies, because ironically (but not surprisingly) these tasks are hard to perform without political trust.  It can be a vicious circle: low levels of political trust inhibit government performance and political institutionalization, which in turn further undermines political trust (Gamson  25 McGraw et al. (2002) point out that trust is conceptually distinct from “suspicion” in the sense that suspicion is a “temporary state that is aroused by specific circumstances and then recedes,” while trust is a more “stable and enduring predisposition” (McGraw, Milton Lodge, and Jones 2002, 362-383).   124 1968; Miller 1974, 951-972; Citrin 1974, 973-988).  This is indeed what is happening in many new democracies.  First, effective government and regime performance is a crucial variable affecting the development of beliefs about regime legitimacy (Diamond 1999).  This includes both economic and political performance.  Although there is no doubt that positive economic performance will be a considerable benefit to the consolidation of democracy, political performance is not any less important.  Furthermore, it should be recognized that the democratic public highly values the responsiveness and accountability of its government. The public in new democracies may value and expect even more from democracy, given that the memories of the authoritarian past likely retain a strong contemporary influence – i.e., for many of them, democracy is not something that they were given but that they fought for. Corruption is also viewed as an important dimension of political performance; but likewise, rampant corruption is a signature characteristic of new democracies with an authoritarian legacy, and it also contributes to a decline of political trust.  Second, and relatedly, political institutionalization is also crucial to the consolidation of democracy.  This includes strengthening the formal representative and governmental structures of democracy (Diamond 1999).  Weak government performance, however, could make political institutionalization even more challenging.  If individuals perceive that the government is not competent in ensuring responsiveness and accountability and is also highly corrupt, they will become more inclined to exercise direct control over the government.  It is noteworthy that the personal experiences of the recent democratic transition, especially ones that are initiated from below, could make this inclination even   125 stronger.  Many citizens in new democracies did experience participation in street demonstrations against their authoritarian regimes and witnessed their power in removing authoritarian leaders from office and successfully leading the country out of authoritarian rule.  When it comes to political institutionalization, however, the experience may not help. When political trust is low, many would prefer to take part in street protests as they did when they fought for democratic transition, rather than to wait for political institutions to work. These attempts will make political institutions even weaker and will make governments scarcely effective. Consequently trust is further undermined.   Indeed, political trust has significantly declined in most new democracies over the past two decades.  As shown in Figure 4.1, while the decline of trust in parliament seems ubiquitous, it is definitely more severe in new democracies.  As indicated by Catterberg and Moreno, trust in parliament fell, on average, 26 percentage points in Latin America, 29 points in former Soviet Union, and 13 points in Eastern Europe between 1981 and 2001 (Catterberg and Moreno 2006, 31-48).  Korea is one of the most dramatic cases, registering a decrease from 67 percent in 1981 under the authoritarian regime, to 34 percent in 1990 right after the democratic transition in 1987. The level of trust dropped even further amidst the Asian Financial Crisis  in 1997 to 10 percent in 2000.  The East Asia Barometer (EAB) in 2003 reveals largely similar outcomes:  only about 15 percent of Koreans express trust in political parties and the parliament, and 27 percent in the national government.  (Zhu 2008).    126 Figure 4.1 Change in political trust (Confidence in parliament) (%) Change in Political Trust between 1981 and 2001 Argentina, -62 Chile , -29 Mexico , -6 Peru    , -5 Belarus , 4 Estonia   , -43 Latvia  , -45 Lithuania, -55 Russia , -25 Ukraine    , -9 Bulgaria   , -22 Czech Republic   , -2 East Germany  , -4 Hungary  , -47 Poland   , -42 Romania  , -2 Slovenia  , -12 Slovakia  , 29 South Africa, -17 South Korea, -57 Turkey, -11 Australia, -25 Belgium, 3 Britain, -4 Canada, -5 Denmark, 11 Finland, -24 France, -8 Italy, 3 Japan, -7 Netherland, 11 Norway, -7 Portugal, 8 Sweden, 5 USA, -14 West Germany, -18 Spain, 0 -80 -60 -40 -20 0 20 40 (%) Change between 1981 and 2001  Source: Adopted from data in Table 1 Confidence in parliament: Development over time in new and established democracies, which is based on 1981 to 2001 World Value Surveys.  Percentages show those saying they have ‘a great deal’ or ‘quite a lot of’ confidence in parliament (Catterberg and Moreno 2006, 31-48).    My argument thus far is twofold: (1) I suggested that political trust has causal importance in trade policy preferences; and (2) I also showed that political trust has significantly declined in new democracies.  In the next sections, by showing how Korean progressive news media engage political distrust to argue against Korea-U.S FTA, I will demonstrate that public opposition to government’s trade liberalization initiatives in large part reflects public condemnation of the government responsiveness and effectiveness (capacity).  Consequently, this indicates that other things being equal, new democracies likely face stronger opposition to any trade liberalization initiatives than advanced democracies.    127 4.4 Korean public opinion on KORUS FTA: The hypotheses  The United States and South Korea signed the Korea-the United States Free Trade Agreement (KORUS FTA) on June 30, 2007.  It was an outcome of ten months of contentious negotiations, which began on June 5, 2006 and were concluded on April 1, 2007. The KORUS FTA is the U.S.’s largest bilateral trade initiative since the North American Free Trade Agreement (NAFTA)26.  For Korea, the FTA with U.S., the world’s largest economy, is even more significant, as it is by far the its largest FTA.  The FTA’s economic impact on Korea is expected to be much greater than its impact on the U.S. economy, since Korean economy is smaller (the world’s 11th largest), more protected, and more dependent on trade than is the US economy (Korus report).  Indeed, the FTA attracted substantial media/public attention.  The debates over the FTA negotiations were splashed on front- page headlines across all local newspapers in Korea, which reflects the widespread and intense public reactions to the FTA negotiations.   Figure 2 presents fluctuations in Korean opinion on the FTA for the ten months of the negotiation (June 2006-April 2007).  The fluctuations in part reflect the sensitivity of the issues on the negotiation table at a given time; but in general they display a pattern that the initial high pro-free trade opinion decreases, but the initial low anti-free trade opinion increases, and thereby the two converge as the negotiations proceed.  The survey conducted by the Korea International Trade Association (KITA) in December 2004, about a year before the negotiations were announced, reveals that 75% of Koreans supported the FTA, which dropped significantly once the negotiations began – it bounced back and forth in the  26 Indeed, the United States International Trade Commission (USITC) estimates that the FTA would add $10 billion to $12 billion to annual U.S. Gross Domestic Product and around $10 billion to annual merchandise exports to Korea.   128 middle of the negotiation, but never reached the initial level of support27.  Opposition to the resumption of the U.S. beef imports in 2008 was even more intense.  Although beef issue was not actually part of the formal FTA negotiations, it was discussed in parallel with the negotiations (CRS report) because resuming importation of the U.S. beef was one of the alleged preliminaries28 demanded by the U.S. as prerequisites for the opening of FTA negotiations.  In 2008, the newly elected Korean President Lee Myung-Bak during the U.S.- Korea summit signed an agreement to lift the ban on U.S. beef imports in the hope that that it would remove the obstacle to ratifying the KORUS FTA in the U.S. Congress.  This agreement however triggered what became known as the “beef-protest” in Korea, which engulfed the Lee administration in a crisis that threatened the FTA.  Hundreds of thousands of people took part in mass protests; and amid the protests, President Lee’s approval rating plummeted to under 20% just five months after his landslide election victory in December 2007.  27 The pattern is in fact not new.  The pattern of Canadian opinion on Canada-US FTA in 1988 was very similar.  Although the pro-free trade majority was never reversed in Canada, the initial gap between support and opposition narrowed significantly by the time the negotiations were completed (Johnston 1992).  As Johnston (1992) shows, almost 80 percent of Canadians supported the FTA in April 1984, which weakened significantly upon the beginning of the negotiations in November 1985, and dropped by almost 30 percent by the time the agreement was reached in October 1987.  This pattern might be ubiquitous, but my argument is that if trust is low, then the divergence in opinions between an abstract idea and a concrete policy (and between before and after the negotiation process starts so that the policy is more specified and concrete) is larger. 28 The preliminaries publicly disclosed include (1) “suspending regulation on pharamaceutical product prices”; (2) “easing government regulation of gas emissions from imported US cars”; (3) “resuming importation of US beef”; and (4) “reducing the compulsory quota which requires South Korean cinemas to screen South Korean films from 146 days per year to 73 days” (For the details, see the website of Korean Alliance against Korea-US FTA at http://www.bilaterals.org/spip.php?rubrique140)   129 Figure 4.2 Changes in Korean public opinion on KORUS FTA during the negotiation (June 2006-April 2007)  Korean Public Opinion on KORUS FTA 58.2% 42.6% 31.1% 45.8% 55.4% 47.8% 50.6% 46.8% 51.2% 29.2% 45.4% 49.5% 43.8% 35.3% 44.5% 43.0% 44.5%42.2% 12.7% 12.0% 11.0% 10.5% 9.3% 7.7% 6.4% 8.7% 6.6% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% Ju ne -0 6 Ju ly- 06 Au gu st- 06 Se pt em be r-0 6 O ct ob er- 06 N ov em be r- 06 D ec em be r-0 6 Ja nu ar y- 07 Fe br ua ry -0 7 M arc h- 07 Ap ril -0 7 Support Oppose Don't know  Note: The polls were conducted during the period from June 2006, when the first round of the negotiation (out of total eight rounds of the negotiations plus one final trade minister level talk) was taking place, to April 2007, which was right after the FTA was concluded. Source: Each poll was conducted by a different public opinion research institute.  The Media Research with a sponsorship of Hankook Ilbo conducted the poll in June 3rd, 2006 (N = 1,000); the Korea Research with Munhwa Broadcasting Service (MBC) in July 9th, 2006 (N = 1,000); the Hankook Research with the Ministry of Finance and Economics in September 26th 2006 (N = 1,000); the Korea Times in October 31st in 2006; the Gallup Korea with the FTA Industry Alliance in December 12th in 2006 (N = 1,000); the Media Research with Hankook Ilbo in January 1st, 2006 (N = 1,000); the Hankook Research with Seoul Broadcasting Service (SBS) and Joongang Ilbo in February 15th, 2007 (N = 10,000); and The Media Research with Korea Broadcasting Service (KBS) in April 4th, 2007 (N = 1,000)     Then what explains Korea’s protectionist sentiment, which became more widespread as the negotiations proceeded and eventually peaked at the beef crisis?  That is, why are Koreans, who have displayed a strong consensus around international trade’s positive impact   130 on the national economy29, ambivalent about the FTA?  Before I present my hypothesis, I discuss three alternative explanations: (1) “identity bias” {{263 Krueger, Anne 1989}}; (2) nationalism (and/or anti-American sentiment); and (3) anti-globalization (anti-neoliberalism).  Trade policy is always contentious; it can therefore be argued that the divergence between support for trade in the abstract and scepticism about trade policy in the particular gets to the root of the controversy over substance in FTAs.  The controversy often revolves around “losers” – i.e., domestic industries/sectors who are likely to lose from the FTA.  As the negotiation proceeds, losers are identified – this implies that it is not only those negatively affected by the FTA that become protectionist, as often assumed by economists30. In other words, it is not only economic factors – i.e., actual costs or benefits that the FTA is likely to generate – but also psychological factors that may play a significant role in forming public attitudes toward protectionism.  Conover and Feldman (1986) once wrote, Political scientists picture the (American) public as one populated by individuals who deal with economic information in a sterile manner devoid of feelings.  Yet the accuracy of this description is belied daily by those people we encounter both in our won personal lives and through the media: people who are depressed over unemployment, worried about being laid off, angry over inflation… (Conover and Feldman 1986, 50-78).   Indeed, the mass media today exerts a large amount of influence on people’s perceptions of mass collectives (Mutz 1998)31.  Given that reality, it is not hard to imagine that as trade  29 Among the 14 countries that were asked whether international trade was good or bad for their economy, Korea was one of the countries that expressed the highest levels of approval (79%), along with China (88%), Israel (88%), and Thailand (79%) (CCFR, April 2007). 30 Numerous studies on public opinion have found that self-interest (narrowly defined) often plays little or no role in determining policy preferences.  For example, self-interest fails to influence mass preferences in such policy issues as bussing, health insurance, unemployment programs, the Vietnam War, and affirmative action (reference). 31 For example, Jacobs and Shapiro (1994) found that there is an interesting disjuncture between the public’s overall contentment concerning their personal healthcare and the public’s dissatisfaction with the quality of healthcare available to others.  While a stable 84-89% of Americans report being personally satisfied with the quality of healthcare received from doctors, only half as many agreed that other patients enjoyed high quality treatment (Jacobs and Shapiro 1994, 9-17).  This shows the importance of what Mutz (1998) called “impersonal influence”: the large amount of support for healthcare reform does not come from Americans who are not   131 negotiations attract public attention, people’s attitudes toward the FTA are influenced by their perceptions of others’ attitudes or experiences portrayed by the media.  For instance, as the debate over FTA’s substance becomes more intense, people become more informed about the losers’ identities.  The knowledge about the losers’ identities likely evokes a more sympathetic response from the public toward their plight than it would if their identities are unknown.  This is what Ann Krueger phrases “identity bias” (Krueger 1989): by drawing on Schelling’s distinction (1984) between statistical and individual specific information, Krueger argues that such identity biases may account for why the political process is often biased towards protection despite the well-known gains from trade liberalization.   For Korea, there were clear victims of the FTA, namely farmers.  Agriculture was high on the U.S. agenda and as expected, negotiations pertaining to access to Korea’s domestic markets of agricultural products including rice, beef, and citrus products were among the most contentious (For details, see the CRS report for Congress on KORUS FTA (2007)).  As the largest trade-flow impact of the KORUS FTA on U.S. was expected to occur in the agricultural sector, the U.S. pressed for complete liberalization in agricultural products.  Korea, however, wanted a number of products to be excluded from the negotiations, because agriculture is the country’s most sensitive and most protected sector. The CRS report states that although agriculture accounts for only 3 % of Korean Gross Domestic Production (GDP) and 7.2% of employment, Koreans’ strong cultural ties to rural areas still makes the agriculture sector a formidable political force (p.13, CRS).  More importantly, the income level of farmers’ households is only about 78.2% of the average  adequately covered by health insurance; rather this support for reform is related to individuals’ perception of anonymous mass collectives, which is overwhelmingly negative (Mutz 1998; Jacobs and Shapiro 1994, 9-17).   132 income of an urban household, and about 87.1% of the average income overall.32  Given that public opinion is likely to be shaped by the feelings that it holds toward the social groups they see as the principle victims of the policy (Nelson and Kinder 1996, 1055-1078; Conover 1988, 51; Kinder 1998), the strong/intense opposition to the FTA might be a reflection of public sympathy for the plight of farmers.  This formulates the first alternative hypothesis.   The next hypothesis relates to nationalism.  There is plenty of empirical evidence that nationalism correlates positively with protectionist sentiment (O'Rourke et al. 2001, 157- 206; Mayda and Rodrik 2005, 1393-1430; Scheve and Slaughter 2001, 267-292).  Mayda and Rodrik (2005) found that among non-economic determinants in the form of values, identities, and attachments, high degrees of neighbourhood attachment and nationalism/patriotism are associated with protectionist sentiment (Mayda and Rodrik 2005, 1393-1430). Likewise, O’Rourke and Sinnott (2001) by showing that protectionist attitudes are strongly related to both patriotism and chauvinism, concluded that “nationalist attitudes exercise some autonomous influence and are a significant factor in the genesis of protectionist policy preferences” (O'Rourke et al. 2001, 157-206).   Indeed, nationalism is the way the overseas media understand the public opposition to the FTA and to the related resumption of U.S. beef imports to Korea, which was very intense at times.  The Washington Post editorial33 about Korean protests of US beef describes the Korean reaction as irrational, compounded by nationalism (Editorial,  32 For details, see http://www.agnet.org/situationer/korea.html. 33 For the details, see the editorial from Washington Post, June 14, 2008: “Seoul's Beef Beef: The Bush administration and Congress must rescue free trade with South Korea.”  Accessible online at http://www.washingtonpost.com/wp-dyn/content/article/2008/06/13/AR2008061303223.html.   133 Washington Post, June 14, 2008).  The New York Times34 likewise writes that the beef dispute is the “test of whether their leaders can resist pressure from superpowers like the United States, even if that pressure is legitimate as is the case in the beef dispute” (Choe Sang-Hun, New York Times, June 11, 2008).  Some academics also hold the similar view: Gi-Wook Shin (2010) argues that the protests over US beef reflect an anti-American sentiment originated from what he calls “(national) identity politics” (Shin 2010).  By comparing the beef protest to the sweeping anti-American reaction to the school girl incident in 200235, Shin traces it to (ethnic) nationalism and anti-American sentiment, which developed at critical moments of Korean history, including the Kwangju massacre in 198036. Given the empirical findings that nationalism in general is associated with protectionist sentiment and the presence of strong nationalism in Korea, it might be nationalism and related anti-American sentiment that accounts for the protectionism37.   34 For the details, see the article by Choe Sang-Hoon from the New York Times, June 11, 2008: “Protests in Seoul more about nationalism than U.S. beef.”  Also accessible online at http://www.nytimes.com/2008/06/11/world/asia/11iht-seoul.1.13635643.html. 35 The incident was one in which  two teenage Korean schoolgirls were killed after being run over by a US army armoured vehicle on an off-base training mission.  The incident provoked anti-American sentiment in Korea because the soldiers involved were found not guilty and released by a US military court.  This prompted hundreds of thousands of Koreans to protest against the US-ROK Status of Forces Agreement (SOFA), which stipulates that US military personnel fall under the jurisdiction of US military courts.  The street protest (candlelight vigil) was recorded as the biggest anti-American protests in Korean history (BBC News, December 10, 2002: “US official met by Korean anger”). 36 On May 17, 1980, the military leadership led by Major General Chun Doo-hwan declared martial law and sent paratroopers to Kwangju to quell a growing democratization movement. The crackdown killed at least 144 civilians (this is an official figure, but the exact number of casualties is estimated much higher).  The United States was often blamed for the massacre, because given its operational control over the Korean forces, it is believed that the United States could and should have stopped the junta from dispatching the paratroopers (Shin 2010). 37 The CRS report writes that it is the United States that sets the agenda of the FTA negotiations, and that South Korean officials simply react to U.S. demands.  Given that that was the way the negotiations proceed – though it was mainly because the U.S. has lower and fewer tariffs than Korea, and Korea is more dependent economically on the US than vice versa – it might have given the Korean public an impression that the Korean government was being dragged into U.S. court, which might have strengthened nationalism and anti-American sentiment.   134  Lastly, it can also be argued that the intense Korean protectionism reflects the underlying anti-globalization/anti-neoliberalism as one can see in any country.  While the definition of globalization is highly controversial, I use the term to refer to a process that is restricted to the economic area, because economic issues such as fear of job displacement, threats to the country’s social programs, risk of worsening  poverty and income inequality, all comprise the key elements of anti-globalization rhetoric.  It might be that most Korean citizens had not formed a specific attitude toward the FTA before the negotiations commenced; but as the debate over the FTA intensified, they absorbed the messages of relevant partisan elites and formed their preferences accordingly (Zaller 1992).  If that is the case, one could argue that protectionism is a reflection of anti-globalization/neoliberalism ideology held by many of the Korean left-wing party (Democratic Party) members.   This chapter does not deny that the three factors above – identity bias, nationalism, and anti-globalization – may play a role in forming public attitudes toward protectionism.  I argue, however, that political distrust is a dominant factor accounting for Korean protectionism among all of these factors.  I will demonstrate it in the next section, by examining media discourse of the FTA issues.  Specifically, I hypothesize that Korean opposition to the KORUS FTA can be largely explained by the public’s distrust of the political system’s (and the politicians’) capacity to effectively deal with its trading partner and their ability to provide adequate compensation for potential losers, as well as distrust of the responsiveness of the policy-making process – i.e., transparency of the process and the government’s effort to communicate with the public through listening and persuasion.  As a newly democratized country, Korea lacks such political trust; and this in large part accounts for the intense public opposition to the FTA in 2006-7 and U.S. beef importation in 2008.   135 4.5 Data and method  This chapter examines the sources of Korean protectionism through the prism of the news media.  Admittedly, survey data would be ideal for testing the relevance of the suggested factors in explaining policy preferences at the individual level; and indeed, most existing studies on policy preferences have relied on public opinion polls.  However, survey data that include relevant variables are not always available, especially when the variables are relatively new, such as political trust.  In other words, given that survey questionnaires are usually designed based on existing theories put to test, survey analysis is often more relevant for theory testing rather than for theory building.  Indeed, the relevance of political trust as an independent variable has largely been underestimated (Hetherington and Globetti 2002a, 253-275); and no empirical investigation has yet been made into the consequences of political trust/distrust on support for specific trade policy initiatives.  In this regard, this research resembles a theory-building process.  In addition, and more importantly, the snapshot nature of survey data makes it difficult to grasp changes in collective public opinion on matters of policy.  The key argument of this chapter is that political trust and protectionist sentiment in the aggregate are in a causal relationship.  However, to demonstrate that the relationship is causal, data need to be placed in a context in which policy is framed, debated, and then re-framed.  Media data provide such context.   This chapter utilizes the Korean news media in examining protectionist sentiment related to specific trade policy initiatives, the KORUS FTA and the re-importation of U.S. beef.  The media matter for two different reasons: the media not only reflect but also shape public opinion.  Content analysis of news media coverage thus inevitably raises an endogeneity question.  The causal importance of trust, if seen in the media discourse, could   136 be a reflection of public opinion; but it could also be the result of media framing.  This study is based on the premise that the mass media mirror public opinion; however, it does not rule out the possibility of the media exploiting political distrust among the public to strengthen protectionist sentiment through framing and/or priming.  Even if it is the result of the media’s agenda setting, the relevance of political trust as a source of protectionism will not be any less significant.  It is probably more important that the media engage citizens’ distrust not anything else to argue against the government’s trade policy initiatives.   To examine how the trade issues were portrayed in the mass media, two newspapers – Hangyoreh Shinmun and Kyunghyang Shinmun – were chosen for analysis.  Newspaper articles were selected because newspaper accounts of policy issues are usually more comprehensive than television news programs, radio or Internet media; moreover, Korean newspapers in general have a  high readership, are national in scope,38 and are known to be highly influential in the policy-making processes.  The particular newspapers chosen, Hangyoreh and Kyunghyang, were selected for their representativeness of anti-FTA views.  Korean news media are sharply divided over many policies issues – i.e., one particular new media often reflects only one side of a given issue – and FTA is no doubt one such issue.  Hangyoreh and Kyunghyang are largely considered the most influential progressive newspapers (especially Hangyoreh) in Korea, and are also known to advocate anti-FTA positions.  As this chapter investigates protectionist arguments, pro-FTA newspapers were excluded from analysis.   38 For instance, the New York Times has a weekday circulation of only about 1 million, while Chosun Ilbo has a daily circulation of more than 2 million.  Given the population of Korea (47 million), 2 million is a significant portion (Shin 2010).   137  The news articles were obtained from Korean Integrated Newspaper Database System (KINDS), the most extensive newspaper archiving database in Korea.  The principal data for this study are full text articles on KORUS FTA and U.S. beef: all articles that mentioned “FTA” from January 2006 to April 2007 and “U.S. beef (Miguksan Soigogi)” from April 2008 to June 2008 were examined39.  The two keywords seem sufficient to cover all the relevant articles.  With “FTA” as a keyword, 2301 articles, and with “US beef” 2021 articles were obtained.  Among the articles collected, the main focus of analysis was placed on editorials (99 editorials for FTA, and 95 for US beef) because they are evaluative in nature – i.e., very suitable  for measuring a newspaper’s interpretation and assessment of factual information.  For the purpose of this study, editorials obtained were categorized based on the four hypotheses suggested above: (1) political trust; (2) identity bias; (3) nationalism (and/or anti- American sentiment); and (4) anti-globalization.  Table 4.1 presents the details.  As defined, the category for political trust contains articles arguing about the government responsiveness and effectiveness.  For responsiveness, articles in this category include those criticizing the government for not publicizing or whitewashing details on negotiations (transparency) and/or for not making an enough effort to communicate with the public through listening and persuasion (communication).  For effectiveness, articles include those that cast suspicion on the government’s resources and capacity to work in the country’s best interest and/or to successfully provide adequate policy measures to minimize potential negative effects of the trade policy initiatives.  The category of identity bias includes articles that mainly touch on  39 The particular time periods were selected because the first round of the FTA negotiation took place in June 2006 and the negotiation was completed in April 2007, and the beef issue emerged during President Lee’s visit to the U.S. in April, 2008 and the mass protest against the beef peaked in June the same year.   138 potential victims of the new policy, e.g., the plight of farmers likely to get worse under the new trade policy, consumers exposed to food-safety risks, or the domestic film industry. Articles that discuss substance in the FTA are included in this category because the newspapers that hold a negative view on FTA are expected to focus more on losers rather winners from FTA.  While the category of identity bias covers issues related to domestic social groups (in-groups) implicated in public disputes over the FTA, the category of nationalism (anti-American sentiment) includes articles generating out-group anxiety - anxiety against the U.S. in this case - by addressing the unfairness of the FTA.  Specifically, articles indicating that global asymmetries of power inevitably lead to unfair negotiation outcomes fall into this category.  Lastly, articles addressing the negative impacts of neoliberalism in general, as a way of criticizing the FTA, are included in the category of anti- globalization.  More specifically, articles that raise general concerns about widening income disparities, job losses, privatization, and weakening social programs fall into this category.  Table 4.1 Arguments against KORUS FTA and the U.S. beef (The coding scheme)  Arguments Trust Identity Bias (Substance in FTA) Nationalism (Anti-Americanism) Anti-Globalization Subtopics  Responsiveness: Transparency; Communication; Unilateral decision-making; Democracy  Effectiveness: Competence as negotiators; Rule enforcement; Proper compensation for victims   Agriculture: Farmers’ plight Consumer: Food safety issue; Screen quota; Genetically modified organisms   Unfair trade: Asymmetry of power Sovereignty: Anti-Americanization    Anti-Neoliberalism: Concerns about widening income disparity; Concerns about job loss; Concerns about privatization; Concerns about social programs (especially, about healthcare)  Key words (Italics Korean) Unpye conceal the fact Sotong communicate with the public Milshilhyeopsang closed door negotiation Ilbangjeok Unilateral Minjujuui Democracy Jolsok Rushed Nongmin farmers Anjeonsung Safety Screen quota GMOs Banmi Anti-Americanism Bulgongjung hyeopsang unfair negotiation Apbak put pressure (on the Korean government) Jukwon sovereignty Yangkuekhwa income polarization Shinjayujuui neoliberalism Minyeonghwa privatization      140 4.6 Explaining trade policy preferences  Figure 4.3 presents percentages of editorials that belong to each category.  The majority of editorials on the FTA (60.6%) are in the category of political trust – that is, as hypothesized, the FTA issues were framed in large part along the storylines that raise public suspicion of government responsiveness (39.4%) and government effectiveness (21.2%).  In the case of the U.S. beef issue, the category of political trust comprises an even higher percentage (73.7%) of editorials with 44.2% for responsiveness and 28.4% for effectiveness, by dominating all the other issues of the other categories – identity bias, nationalism, and anti- globalization.  At the beginning of the debates on the FTA (when the FTA first attracted media attention), I found a number of editorials revolving around concerns about the opening of agriculture, the loosening of the screen quotas, and/or neoliberal concerns in general, such as income inequality; but once the negotiations commenced, these concerns were almost muted by criticisms directed at the government for its lack of transparency and responsiveness, and its incompetence.  Likewise, the immediate editorial reaction to the beef deal focused on food safety concerns for consumers and concerns about its impact on Korean livestock industry; but as the debates proceeded, these issues entirely disappeared. Since two weeks after President Lee Myung-bak signed the deal, no single editorial raising concerns about domestic livestock industry or consumer food safety has been found.   141 Figure 4.3 Arguments against the FTA and the U.S. beef import Focus of Editorials on KORUS FTA 39.4% 21.2% 60.6% 6.1% 15.2% 9.1% 10.1% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% Tr us t ( Re sp on siv en es s) Tr us t ( Ef fe ct iv en es s) Tr us t ( Re sp on siv en es s p lu ... Id en tit y b ias  (S ub sta nc e i ... N ati on ali sm  (A nt i-A m er ic. .. An ti- glo ba liz at io n (A nt i-n ... O th er s Arguments P er ce n ta ge  Focus of Editorials on U.S. beef 44.2% 28.4% 73.7% 5.3% 7.4% 5.3% 9.5% 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% Tr us t ( Re sp on siv en es s) Tr us t ( Ef fe ct iv en es s) Tr us t ( Re sp on siv en es s p lu ... Id en tit y b ias  (S ub sta nc e i ... N ati on ali sm  (A nt i-A m er ic. .. An ti- glo ba liz at io n (A nt i-n ... O th er s Arguments P er ce n ta ge       142   More specifically, analysis of the editorials on the KORUS FTA reveals that anti- FTA arguments during the negotiations revolved around the following criticisms directed at the government:   The government (Roh administration) had done little preparatory work needed to get the best deal for the country out of the negotiation40;   (Immediately after the first round of the negotiation concluded) The government refused to release documents that may have revealed controversial details on the agreement with the US – without any checks and balances, the public’s future is completely at the mercy of the negotiators;   While the government stakes everything on the completion of the FTA within a given time period (as if conclusion of the FTA is by itself an objective), it is rushing to sign the FTA without consulting the National Assembly or holding public hearings.   That trade negotiations proceed without the National Assembly (the legislative body of the government41) being fully informed, which is not in accordance with the principle of democracy;   The government deliberately misled the public – although the government denied it, it turned out that there were a number of issues agreed to under the table – e.g., the US-Korea Understanding on Agricultural Biotechnology was negotiated on the sidelines of the FTA;   The government blocking protests against the FTA is no different from Korean returning to dictatorship of the 1980s. Interestingly, the editorials on the U.S. beef issue show almost the same pattern.  While criticisms of the government regarding the beef issue were more intense than those regarding the FTA, as one can see from the number of editorials at a given period of time,  40 In fact, the official launch of negotiations for KORUS FTA came as a surprise to many Koreans.  As shown in the term “left-wing neoliberalism,” in President Roh’s words, the government’s drive for the FTA was viewed as a puzzling and abrupt move (Lim, 2006). 41 Unlike in the US, the Korean National Assembly, the legislative body of the Government, does not have authority for trade negotiations and, therefore, is not directly involved in FTA negotiations (Cheong and Cho, 2009).   143 arguments against re-importation of U.S. beef are also centered on scepticism and suspicions of the government dealing with the issue and evolved into an argument about a crisis of representative democracy.  The main arguments can be summarized as follows:   The government rushed to end the beef deal with the U.S. in consideration of the planned summit talks at Camp David;   The government failed to provide the public with any kind of persuasive information and relevant documentation.  It is irresponsible for the government to open up the market to U.S. beef (including beef from cattle over 30 months of age and older which is generally not sold for food consumption in the U.S. and elsewhere)42 while just saying to people, “if you don’t like it, don’t buy it” (President Lee’s speech delivered to the public on April 22, 2008)   The government is not competent to deal with the problem of enforcing the beef labelling regulations, which is required to secure consumers’ rights to choose products (May 6, 2008, Kyunghyang);   The government provided an explanation to the public that the decision to re-open the market to the U.S. beef was made in accordance with the recent U.S. introduction of measures to tighten regulations on animal feed, which, however turned out to be groundless.  The US government in fact did not tighten these regulations, but actually loosened them.  Thus, either the government made a serious mistake during the negotiations, or it is telling a lie.  “It is very hard to trust the government assertion that Americans and Koreans will be eating the same beef” (May 15, 2008, Hangyoreh);   This is a crisis of representative democracy.  The government has to listen to the public, as “the republic of Korea is a democratic republic and the sovereignty of the republic of Korea resides in the people and all state authority emanates from the people” (Article One of the Constitution).  If the president goes against the will of the people, the direct democracy – i.e., politics in the street – will continue (June 10, 2008, Hangyoreh).    Given that the FTA and the beef deal were signed under the two different political leaderships – the left-wing leadership and the right-wing leadership, respectively – it is  42 The controversy over the beef issue was centered on the government decision to include the import of beef from cattle over 30 months old and to allow a lower standard of specified risk materials (SRM) in their initial agreement with the US than even the US allows.   144 interesting that the arguments against the FTA differ little from the arguments against the US beef.  Scepticism and suspicions of government institutions lie at the core of the arguments against the two specific trade initiatives.  The fact that the pattern of arguments did not respond to changes in partisan control of the government suggests that political distrust in Korea is not a mere reflection of political leaders, and rather, it results from negative evaluations of government institutions in general.  4.7 Conclusion  Political trust has causal significance in forming trade policy preferences.  The results above demonstrate that public opposition to a government’s trade liberalization initiatives in large part reflects the public condemnation of responsiveness and effectiveness of government institutions.  This further confirms the assertion that new policy initiatives that inevitably involve uncertainties and risks, such as trade policies, require political trust, since fears of uncertainties and risks can only be eased by trust.  Low levels of political trust are a common characteristic of many new democracies. In this regard, the results above are suggestive of the need for further study.  The causal significance of political trust in trade policy preferences is likely to be generalizable to most new democracies that are characterized by weak political institutions and ineffective regime performance.  It is noteworthy that the majority of the criticisms of the two trade initiatives in Korea were directed at the government (the President and the negotiators), and that the protectionist arguments evolved into arguments about a crisis of representative democracy. Apparently, under the Korean Constitution, the National Assembly – the legislative body of the government of Korea – does not have authority for trade negotiations.  As the legislative   145 body is not allowed to be directly involved in FTA negotiations, the mediating political institutions that translate the public preferences into trade deals were largely absent.  If the public feels that the policymaking process is not responsive and transparent enough to make their voices heard, and that government institutions in charge are too incapable, ineffective, and inefficient to carry out their responsibilities, just as the Korean cases illustrated, then the public should distrust and reject its policies, especially those which involve a great deal of uncertainty and risk.  In short, political distrust is a powerful cause of protectionism in Korea; and the casual relevance of political trust in trade policy preferences in other new democracies invites further research.       146 5 DOES DEMOCRACY ENCOURAGE TRADE LIBERALIZATION? FACTOR ENDOWMENTS AND PROTECTIONIST SENTIMENT IN LESS-DEVELOPED COUNTRIES  5.1 Introduction  Does democracy encourage trade liberalization?  The effects of regime types on trade policy have been a subject of a scholarly debate amid the seemingly concurrent processes of the two macro-phenomena – namely, democratization and globalization – in recent years. Empirical work testing for the impact of democracy on the liberalization of trade policy has largely concluded that democracy positively influences economic openness (Tavares 2008, 163-168; Milner and Kubota 2005, 107-143; Milner and Mukherjee 2009, 163-181; Kono 2008, 942-955; Kevin O'Rourke 2007; Eichengreen and Leblang 2008, 289-334).  Theoretical analyses of the causal impact of democracy on economic openness are however relatively underdeveloped (Milner and Mukherjee 2009, 163-181).  Empirical work that is mostly large- n often adopts the Hecksher-Ohlin (and Stolper-Samuelson) logic to provide causal stories behind the findings.  According to this logic, which has been advanced by Mayer (1984) and Yang (1995), the emergence of democracy fosters trade openness in less developed countries (LDCs) because democratization in LDCs will broaden the franchise to the wider population, most of whom are low-skilled workers that the H-O predicts are most likely to gain from trade liberalization (Mayer 1984, 970-985; Yang 1995, 956-963).  The goal of this chapter is to test the relevance of the H-O as an underlying logic of the causal relationship between democracy and trade liberalization.  Though useful, the   147 causal stories based on the H-O logic have been subject to criticism for its over- simplification, by treating labour in LDCs as having trade preferences determined by the country’s relative factor endowments.  By tackling this issue, this paper re-explores the causal link between democracy and trade.  More specifically, the paper attempts to provide empirical answers to the following questions.  First, do the electorates in LDCs indeed more likely prefer lower levels of protection as predicted by H-O?  In other words, are individual economic gains and losses a primary determinant of their preferences over trade liberalization?  Second and related, if not factor proportions, i.e., distributional implications of trade, what explains the considerable degree of national variation across countries in protectionist sentiment?   First and foremost, this paper will provide an opportunity to rethink the relationship between democracy and trade openness by demonstrating that empirical support for the predicted effect of the H-O is weak.  In fact, while it has often been suggested that democracy affects trade conditional on factor proportions – the capital-labour ratio in particular – a causal mechanism linking factor endowments to trade preferences has not yet been established.  The so-called “Human Capital Hypothesis (HCH),” which claims that trade preferences stem from individuals’ earning power in the labour market, namely “human capital,” for instance, generates contradictory predictions to the factor-endowment based models (Baker 2003, 423-455).  In the H-O setting, trade liberalization brings losses to owners of scarce factors and gains to owners of the abundant factors, thereby affecting individual trade preferences.  The HCH, however, assumes that trade liberalization affects trade preferences by increasing labour market instability and insecurity – i.e., the hypothesis assumes that regardless of the country’s factor endowment proportions, the low-skilled are   148 always more vulnerable than the high-skilled to the shifting configuration of the labour demand, thereby tending to be more protectionist (Baker 2003, 423-455; Rodrik 1997; Scheve and Slaughter 2004, 662-674).  Indeed, the political economy literature on globalization has pointed out that trade liberalization not only generates winners and losers depending on the country’s relative factor endowments, but also increases worker insecurity due to increasing the wage-elasticity of demand for labour.  Basically, both H-O and HCH, though generating contradictory predictions to each other, seem plausible.  The study therefore puts the two logics to the test, and demonstrates that the HCH in fact works better than H-O.  Taking this argument further, this chapter also provides new evidence on non- economic determinants of individual preferences over trade liberalization.  As the main independent variables, I concentrate on country-level differences in attitudes toward income inequality and national pride.  Scholars in fact have shown that non-economic determinants in the form of values, identities, and attachments also play an important role in explaining the variation in protectionist sentiment (Mayda and Rodrick 2005, 1393; O'Rourke et al. 2001, 157-206).  O’Rourke and Sinnott (2001) find that both patriotism and chauvinism are powerful determinants of protectionist sentiment, which is later confirmed by Mayda and Rodrik (2005).  Other than nationalism, however, there has not been much discussion concerning non-economic determinants of trade preferences; and moreover, no systematic research has been conducted to examine if country-level differences in values and identities have an impact on shaping trade attitudes.  In this chapter, I will show that not only nationalism but also attitudes toward income inequality are associated with trade preferences;   149 and that such values vary greatly among countries, which consequently in large part explains the cross-country variation in trade preferences.  By investigating the impact of cross-national differences in relative factor endowments and values and identities on individuals’ attitudes toward trade liberalization, this study seeks to make methodological, as well as substantive and theoretical, points.  In fact, most of the studies exploring individual preferences over trade liberalization thus far have relied on single-level analysis43.  This is surprising given that the H-O model on which these studies are based suggests that trade attitudes are largely a function of factor proportions of a given country, which is a contextual, country level factor.  This chapter focuses on country-level characteristics – such as factor endowment proportions (capital- labour ratio), economic openness, and country average levels of nationalist sentiments and attitudes toward inequality – the degree of which vary across countries as determinants of trade preferences, while acknowledging that the inclusion of country-level variables in the model necessitates multilevel modeling techniques to account for lower variance at the individual level due to individuals being nested within countries.  Basically, this study will provide a test of the generalizability of findings across countries by employing multilevel analysis that accounts for both individual- and country-level variation in estimating country- level regression coefficients (Steenbergen and Jones 2002, 218-237).  One thing to be noted is that this study is not arguing that it is necessarily wrong that democracy encourages trade openness.  Nor is it trying to provide an alternative causal mechanism that links democracy to pro-trade policy.  Rather, this study aims to test the  43 Mayda and Rodrik’s paper has a multilevel components – needs to elaborate it further.   150 relevance of the H-O model as a theoretical underpinning of the argument presented by Milner and Kubota (2005), O’Rourke and Taylor (2006), and Tavares (2006).  By employing multilevel analysis, I demonstrate that it is empirically groundless that the electorates in LCDs – labour-abundant countries by definition – favour trade openness, and that newly democratized governments in LCDs thus have incentives to lower trade barriers.  The investigation of non-economic determinants of trade preferences will provide confirmation that the H-O based claim that democracy fosters trade liberalization is weak: to the extent that protectionist sentiments are tied up with deep-seated social values such as national identity or attitudes toward inequality, distributional implications of trade liberalization are unlikely to shape trade attitudes – that is, the H-O logic is unlikely to operate.  5.2 Theories and hypotheses  5.2.1 Implications of public opinion studies for trade preferences  Does democratization (or the level of democracy) affect trade liberalization?  The political economy literature on the impact of political regimes on trade liberalization has revealed that democracy positively influences trade liberalization (Tavares 2008, 163-168; Milner and Kubota 2005, 107-143; Milner and Mukherjee 2009, 163-181; Kevin O'Rourke 2007).  The causal argument, though backed up by empirical research, is theoretically underdeveloped. The theoretical story behind the empirical work is in common built on the Heckscher-Ohlin (and Stopler-Samuelson) model, which posits that trade liberalization has an asymmetric impact on capital-abundant and labour-abundant countries.  According to the model, as countries integrate with world markets, the low-skilled labour and the poor in LDCs tend to favour more trade openness, whereas those in developed countries tend to be more   151 protectionists.  This is because the model predicts that the low-skilled tend to gain from trade liberalization in LDCs where labour is abundant (relative to capital) through increased wage and reduced consumer prices, and the reverse is true in developed countries where labour is scarce.  The H-O model, combined with Duncan Black’s (1948) median voter theorem44, seems to support the argument that democratization positively influences trade liberalization.  Given that democratization by definition implies transferring power from non-elected elites to the wider population, and that most of the electorate – and notably, the median voter – are relatively capital poor in LDCs thus being more in favour of trade, democratization is likely lead to more economic openness (Milner and Kubota 2005, 107- 143; Milner and Mukherjee 2009, 163-181; Milner and Mukherjee 2009, 163-181; Mayer 1984, 970-985; Yang 1995, 956-963; Kono 2008, 942-955).  In short, the theoretical argument about democracy and trade policy is built on the assumption that the electorate prefer more trade openness in LDCs, and that democracies are more responsive than autocracies to electoral pressures45.   Is it realistic to assume that the electorates in LDCs are in favour of trade liberalization?  This assumption is critical, because if the assumption does not hold and hence the electorate were generally protectionist, then democracy would have the opposite  44 The median-voter theorem posits that under the condition of a uni-dimensional policy space, single-peaked policy preferences, and a majoritarian rule, the policy chosen by the median voter cannot be dominated by any other alternative (Black 1958). 45 The assumption that democracies are more responsive than autocracies to popular pressures is also questionable, although in this chapter I mainly tackle the first assumption that the electorate in LDC are in favour of trade liberalization.  Indeed, trade policy scholars have indicated that unlike small producer groups, mass publics often face collective action problem (Olson 1971; Gilligan 1997); and they might also argue that public ignorance about trade policy is said to be “rational” because the cost of educating oneself about the policy to make an informed decision likely outweighs any potential benefit from that decision.  Given both collective action problem and rational ignorance, claiming that democracy encourages trade openness on the ground that democracies are more responsive than autocracies to popular pressures seems problematic.  In order for democracy’s responsiveness to operate in a way to foster trade openness, one has to demonstrate first that there is indeed public pressure for more openness in LDCs.   152 effect.  The point is that even if it is empirically true that workers and the poor in LDCs gain from trade liberalization as predicted by H-O, it is implausible to assume that trade preferences of the electorate are dictated by material benefits they would accrue from trade. The literature on public opinion studies is replete with evidence that immediate and tangible self-interest that individuals see at stake in policy issues plays surprisingly little or no role in determining policy preferences.  Self-interest fails to influence mass preferences in such policy issues as bussing, health insurance, unemployment program, the Vietnam War, and affirmative action (Sears et al. 1980, 670-684; Lau and Sears 1981, 279-302; Kinder and Sanders 1987; Kluegel and Smith 1982, 518-532; Kinder 1986, 151-171; Sears and Lau 1983, 223-252; Sears and Kinder 1985, 1141-1147).  When faced with affirmative action, white and black Americans formulate their views without calculating personal costs or benefits (Kluegel and Smith 1986; Kinder and Sanders 1996).  The literature on sociotropic voting tells us that individual voters tend to respond to a nation’s overall economic condition rather than their own pocketbook (Kinder and Kiewiet 1981, 129-161; Mutz 1992, 89-122; MacKuen, Erikson, and Stimson 1992, 597).  Substantial research in economic policy domains as well has demonstrated that self-interest rarely shapes the formation of policy opinions: for example, personal economic grievances, such as unemployment, fail to have any direct impact on forming opinions for policies designed to alleviate economic distress (Schlozman and Verba 1979).   In the literature on the political economy of trade, however, such psychological perspectives have been largely disregarded; political economists have been hesitant to sacrifice the rationality assumption in a narrow economic sense (“pure” self-interest). Individuals are often assumed to be rational information processors who carefully review the   153 available information about trade policy, and then make political judgments based on acquired knowledge and its relation to their personal economic conditions.  Evidence, however, has shown that such “cognitive” assessments of economic information only partially explain individuals’ economic policy evaluations; and that emotional reactions to policy issues are in fact ubiquitous (Conover and Feldman 1986, 50-78).  The case of trade policies is not the exception: Fernandez and Rodrik demonstrate that even in cases which trade reform would prove quite popular ex post, uncertainty regarding the distribution of gains and losses from reform may generate a bias among the mass public toward the status quo (and hence against trade liberalization) (Fernandez and Rodrik 1991, 1146-1155). Moreover, the costs and benefits of trade liberalization are not easy for laypersons to identify in advance as economic interdependence between states has become more complicated; and citizens often have difficult time understanding the connection between their personal economic well-being and policy issues.  If that is often the case, citizens may need to rely on information shortcuts – expert advice, ideology, party cues, and campaigns – and thereby become susceptible to issue frames provided by political elites.  If affective components, such as anti-globalization, are added to elite messages, citizens may have strong attitudes toward the policies, even if those policies are not expected to have a direct impact on these citizens themselves.  In short, a substantial number of public opinion studies suggest that it may not be exceptional that the electorate that will either benefit from trade or be neutrally affected might take a strong protectionist position, or vice versa.   I hypothesize that the electorate in LDCs are not necessarily in favour of trade liberalization – rather, they are more likely to be protectionist than those in developed countries – for the following reasons.  First, economic insecurity – job insecurity and labour   154 market insecurity – has been considered the most frequent and powerful rationale for protectionism (Irwin 2005a; Hays, Ehrlich, and Peinhardt 2005, 473; Garrett and Mitchell 2001, 145-177).  Indeed, the welfare-state literature has argued that economic integration increases economic insecurity among workers thereby generating demand for more generous social insurance that compensates workers for a riskier environment (Hays, Ehrlich, and Peinhardt 2005, 473; Garrett 1998; Burgoon 2001, 509-551; Scheve and Slaughter 2004, 662- 674).  If more open economies have greater exposure to the risks emanating from turbulence in global economy, then what determines trade preferences is probably not so much relative factor endowments as individuals’ earning power in the labour market, namely, “human capital” (Baker 2003, 423-455).  As the Human Capital Hypothesis postulates, low-skilled individuals are in general more vulnerable than high-skilled individuals to the shifting configuration of the labour demand since they tend to be less adaptive to the new environment generated by trade liberalization.  The hypothesis has been empirically supported: the literature of political economy of trade has shown that trade liberalization increases worker insecurity in the labour market (Rodrik 1997; Rodrik 1998, 997-1032; Scheve and Slaughter 2004, 662-674).  Scheve and Slaughter (2004), for example, find that workers care about the risk of unemployment and the volatility of earnings, and that foreign direct investment (FDI) by multinational enterprises (MNEs) is positively correlated with individual perceptions of economic insecurity.  In short, if it is human capital rather than relative factor endowment that matters in determining trade preferences, it is questionable that the low-skilled are in favour of trade liberalization not only in developed countries but also in LDCs.    155 Second and related, numerous studies on public opinion have found that individuals tend to formulate policy preferences based on their perceptions of collective conditions – e.g., perceptions of how a given policy affects a country as a whole – rather than their personal problems and achievement (Kinder and Kiewiet 1981, 129-161; Mutz 1998; Kinder and Kiewiet 1979, 495; Mansfield and Mutz 2009a, 425-457).  The empirical evidence has been supportive of the so-called “sociotropic hypothesis” in many different policy areas, and trade issues are not the exception.  Mansfield and Mutz (2009), for example, have recently demonstrated that trade preferences as well as attitudes about other domestic economic domains such as unemployment are guided less by narrowly defined self-interest than by perceptions of how trade policy affects the country collectively (Mansfield and Mutz 2009a, 425-457).  It is important to note that the sociotropic hypothesis provides a fundamentally different causal mechanism linking economic conditions to trade preferences from what has been suggested by economic trade theories.  It is basically an information-based explanation (Mutz 1998; Mansfield and Mutz 2009a, 425-457), viewing individuals as bankers rather as peasants46 (MacKuen, Erikson, and Stimson 1992, 597).  By suggesting that individuals respond to mass-mediated information beyond their personal experience, the hypothesis provides the potential for country-level contextual factors, including country-level differences in relative income inequalities, perceived social injustice, and confidence in the working of domestic political institutions47, to influence the formulation of policy preferences.  By political institutions here, I mean the set of institutions governing political decision-making.  Political scientists have long been interested in enhancing capacity of  46 Mackuen et al. (1992) have found that American voters tend to make a political decision based on an abstract view of national conditions or expectations of national conditions in the future (voters as “bankers”), rather than based on assessments of their current economic conditions (voters as “peasants”) (MacKuen, Erikson, and Stimson 1992, 597). 47 I discussed in the earlier chapters that confidence in political institutions likely influences public attitudes toward trade liberalization.   156 political institutions to enable the state to credibly commit to preserving markets – i.e., the capacity to provide a secure and predictable political foundation for the markets (Williamson 1993, 453; Williamson 1985; North and Weingast 1989, 803-832; Weingast 1995, 1).  As pointed out by the welfare-state literature, markets that societies do not recognize as legitimate cannot be sustainable; and hence that it is important to give people the confidence that the risks of trade liberalization will be shared (Abdelal and Ruggie 2009; Polanyi 2001). In the case of newly democratized LDCs, however, politicians often have difficulty in making such credible promises (Keefer and Vlaicu 2008, 371); and citizens are exposed to illegitimate and corrupt practices of government on a regular basis.  The point is, if it is sociotropic insecurity that shapes trade preferences, there is little reason to believe that the electorate in LDCs characterized by weak domestic political institutions are in favour of trade liberalization, a market-oriented policy.  5.2.2 The factor endowments model and trade preferences  Three recent papers, Milner and Kubota (2005), O’Rourke and Taylor (2006), and Tavares (2008), have used aggregate data to address the relationship between democracy, factor proportions, and trade liberalization and confirm that deepening democracy leads to trade liberalization in labour-rich countries (Tavares 2008, 163-168; Milner and Kubota 2005, 107- 143; O'Rourke and Taylor 2006).  Milner and Kubota (2005) present evidence for LCDs suggesting that average tariff rate depends on the interaction of the country’s factor endowments and the levels of democracy.  O’Rourke and Tayler (2006) likewise test the political implication of the H-O model, and confirm that democracy is associated with lower tariff in LDCs where workers stand to gain from trade liberalization.  Taveres (2008) also   157 tests and confirms that deepening democracy is associated with demands for increased trade openness by using aggregate, country-level data.   The aggregate, country-level analysis, however, is not sufficient to back up the theoretical relevance of the H-O model – that is, the empirical evidence at the aggregate level tells us little about causal stories for the relationship between democracy and trade policy. Milner and Kubota (2005), O’Rourke and Tayler (2006), and Taveres (2008) all hypothesize that democracy has a positive effect on trade liberalization in LDCs because democratization politically empowers the abundant factor – i.e., workers and the poor – who prefer free trade. The hypothesis is however based upon two untested assumptions: (1) public opinion (the opinion of the median voter) affects trade policy; and (2) trade attitudes are largely a function of who is personally benefited or hurt by trade in labour market, and therefore workers and the poor (the median voter) in LDCs support free trade.  First, it is debatable that public opinion does indeed affect trade policy; and if not, i.e., mass publics are politically ineffectual, then the H-O model does not any longer serve as a theoretical underpinning of the apparent positive relationship between democracy and trade policy.  In fact, trade policy scholars traditionally have assumed that public opinion is unimportant because mass publics unlike interest groups inevitably face collective action problems (Olson 1971; Gilligan 1997). Second, it is not yet clear if electorates in LDCs indeed want open trade; and if not, the H-O model will suggest that democracy have a negative impact on a trade policy.  Although the two assumptions are equally important, this chapter mainly tackles the second assumption by empirically testing whether electorates in LDCs are indeed more favourable toward trade liberalization than those in developed countries, as predicted.    158  In fact, empirical studies of individual trade policy attitudes have generated a fair amount of support for the H-O model, which seems to provide some empirical confirmation for the political implication of the H-O model used by scholars above.  By using cross-national survey data, O’Rourke and Sinnott (2001) and Mayda and Rodrik (2005) demonstrate that individual skills correlate negatively with protectionist sentiment in developed countries but the correlation weakens and even reverses in LDCs.  In a Canadian context, Balistreri (1997) and Beaulieu (2002) also conclude that Canadian attitudes toward Canada-U.S. Free Trade Agreement of 1988 are largely predicted from the H-O model (Beaulieu 2002, 99-131; Balistreri 1997, 1-17); and Scheve and Slaughter (2001) examine the U.S. National Election Studies surveys in 1992 and 1996, and also find that a class characteristic (based on skill-levels) is a good predictor of trade opinions among American public (Scheve and Slaughter 2001, 267).  The empirical patterns found here have generally been understood as corresponding to the H-O model.   The individual-level evidence is not without exceptions, however.  Baker (2003 and 2005) for example counters the H-O model, by revealing that the low-skilled and the poor tend to be more protectionist than the skilled in Latin American, i.e., labour abundant countries (Baker 2003, 423-455; Baker 2005, 924).  Alternatively, he suggests a consumption- based model: in developing countries like Latin America, which practiced long decades of import substitution, individuals are more likely to access trade policies based on their visible impact on consumer choices, rather than on their impact on wages or job opportunities. Scholars like Hainmueller and Hiscox (2006, 2007) also provide an alternative plausible explanation for the causal effect of skill-level on attitudes toward trade liberalization (Hainmueller and Hiscox 2006, 469-498; Hainmueller and Hiscox 2007, 399-442)  While   159 questioning that the relationship between skill levels and trade attitudes is a primarily a product of distributional concerns about the labour market and wage as typically assumed by trade theories, Hainmueller and Hiscox argue that the relationship may simply reflect the impact of education (college education or economic literacy) on attitudes toward trade48.  In short, there have been considerable discussions and empirical tests of the H-O model employing individual-level data; but the empirical results are mixed.  It is not yet clear whether the low-skilled in LDCs are in favour of trade liberalization; and even if it does, it is also not clear whether individual gains and losses in the job market are a primary determinant of their preferences over trade liberalization.   In fact, as shown above, research focused at the macro level – i.e., research on the relationship between regime types and trade openness – and research into the attitudes of individuals – i.e., survey analysis testing the H-O – have been conducted largely in separate worlds.  This is problematic given that the latter has served as a theoretical foundation of the former.  By tackling this issue, this study attempts to integrate what we know about individual differences in opinions and socio-economic status into what we know about contextual, country-level differences.  By combining the two levels of analysis in a single comprehensive model, this study will be able to directly explore how individuals’ attitudes  48 Hainmueller and Hiscox (2006) argue that exposure to a college education that teaches “economic ideas about the overall efficiency gains for the national economy associated with greater trade openness (economic literacy)” is critical contributor to the positive relationship between skill-level and support for free trade in the United States (Hainmueller and Hiscox 2006, 469-498).  According to Hainmueller and Hiscox, this argument is supported by two findings: (1) the positive relationship between skill level and pro-trade sentiment is shown almost identically among people in the active labour force and those who are not; and (2) college-educated individuals support trade liberalization than others, whereas other types of education have no significant effects on attitudes (the so-called “the college plateau effect”).  Although it is questionable whether these two findings indeed demonstrate their argument, Hainmueller and Hiscox’s alternative explanation is worth noting in the sense that it suggests costs and benefits of trade may not be a primary factor that shapes mass preferences over trade liberalization.   160 toward trade are shaped by and embedded in variable-contextual parameters, e.g., a relative factor endowment of a given country.  5.3 Data and methods  In order to assess the relation between factor proportions and trade attitudes, which is the focus of this study, we need measures of these two phenomena for a large number of countries.  As no single dataset offers both levels of information, various data sources need to be combined.  For the attitudinal indicators and sociodemographic control variables I use the Asia Europe Survey (ASES) round 2000-2001, fielded in eighteen Asian and European countries: Japan, Korea, China, Taiwan, Singapore, Malaysia, Indonesia, Philippines, Vietnam, the United Kingdom, Ireland, Germany, Sweden, France, Spain, Italy, Portugal, and Greece. The ASES data are well suited for this analysis because the data include countries that vary in capital-abundance and regime types, which are the important macro-level variables for this study.  For country-level data, this study uses Penn World Table (PWT), collected by Center for International Comparisons of Production, Income and Prices at the University of Pennsylvania (Version 6.3, updated August 2009).  PWT includes purchasing power parity and national income accounts converted to international prices for 188 countries including the 18 countries for this study49.   In this study, trade attitude at the individual level is used as a dependent variable.  In order to measure individual trade attitude, I focus on survey responses to the following question: “Please tell me how much you agree or disagree with the following statement: [Country] should limit the import of foreign products.”  After deleting the “Don’t know”  49 For more detailed information about PWT, see http://pwt.econ.upenn.edu/aboutpwt2.html.   161 and “NA, refused” responses, I transformed survey questions into the ordinal dependent variable Protectionism on a scale from 1 to 5.  I assigned a value of 1 to respondents who answered “strongly disagree” and a 5 to those who answered “strongly agree”.  Higher values of Protectionism thus correspond to more protectionist sentiment.  I also created a binary variable labelled Protectionism dummy.  Protectionism dummy is set equal to 1 for individuals opposing import of foreign product, and to 0 for the rest.  Figure 5.1 Protectionist sentiment by country  Notes: The graph presents the mean scores per country on protectionist sentiment.  Standard deviations are as follows: Singapore 1.035953, Germany 1.159159, Sweden 1.346925, Japan 0.889645, Taiwan 1.073490, China 1.223639, Spain 1.141654, France 1.332411, the United Kingdom 1.157257, Italy 1.305793, Ireland 1.174243, Portugal 1.255794, Indonesia 1.054262, Malaysia 1.041937, Philippines 1.205915, South Korea 1.087926, Thailand 1.026285, and Greece 1.015761.  Countries, Real Gross Domestic Product per Capita   162 (G-K method, current price) of which are less than 60 percent relative to the United States are categorized as Less-Developed Countries (LDCs). Sources: The Asia Europe Survey (ASES) 2000-01 and the Penn World Table (PWT) 2001.    I am primarily interested in the effect of factor proportions of a country on trade attitudes, thereby testing the H-O based claim that the electorates favour trade liberalization in capital-scarce countries (LDCs) and are against it in capital-abundant countries (developed countries).  Before I present a series of multilevel models exploring the association between factor proportions (measured by GDP per capita) and protectionist sentiment, as a first pass through the data it is instructive to look at summary statistics of protectionist sentiment by country.  I plotted average protectionist sentiment by country in Figure 5.1, while highlighting the LDCs with a red colour.  There is indeed a considerable degree of national variation across countries.  Figure 5.1 shows that countries vary greatly in terms of protectionist sentiment; and that unlike the H-O predictions, most of the LDCs are among those highly protectionist countries.  Countries, real Gross Domestic Product (GDP) per Capita in 2001 (G-K method, current price) of which is less than 60 percent relative to the United States50 are categorized as LDCs; and they are Greece, Thailand, South Korea, Philippines, Malaysia, Indonesia, Portugal, China, and Taiwan.  With an exception of the two countries – China and Taiwan – all the LDC countries display stronger protectionist sentiment than most of the developed countries in the region.  A more systematic examination of the relationship between factor proportions and protectionist sentiment will be followed; but it is noteworthy that even a simple dotplot reveals that the underpinning of the argument that democracy encourages trade openness is not empirically supported. Basically, it is far from the empirical truth that electorates in LDCs want open trade.  50 When I use GDP per capita as a criterion, and categorize those, GDP per capita of which is less than USD 20,000 into LDCs, the countries included in the group of LDCs are exactly the same as above.   163   To produce a more systematic analysis, the combined dataset including the ASES survey and the PWT data will be analyzed using random intercept (and random coefficient) multilevel models.  The dataset is therefore nested data where respondents are sampled within different national contexts.  In fact, multilevel data is common in this research area – for example, the datasets such as the International Social Survey Programme (ISSP) and the World Value Survey (WVS), which are commonly used to test if individual trade preferences are correlated with an individual’s skill levels in manner predicted by the H-O, have a multilevel characteristics: the datasets cover a large sample of individuals from a number of different countries.  Despite the prevalence of cross-sectional data, a multilevel modeling technique has not been widely used, however (Steenbergen and Jones 2002, 218-237).  As Steenbergen (2002) nicely summarizes, ignoring the multilevel character of data while treating them as having N independent observations invites errors, because given that individual-level factors are likely influenced by country-level factors, using pooled data violates the assumption that the errors are independent.  When testing the H-O implication, both O’Rourke and Sinnott (2001) and Mayda and Rodrik (2005) conducted individual-level regressions with country (region) indicators, followed by simple country-level regression of the estimated country effects.   This is slightly more elaborate analysis proceeded in two steps.  They ran a series of logit and probit regressions on individual countries with trade attitudes as a dependent variable, which generated separate predictors for each country.  Then they perform a simple linear regression at the country level, considering the estimated coefficients of the country indicators as the new data.  The two-step analysis seemed reasonable; and in fact they managed to discover   164 there is a strong and tight relationship between a country’s factor endowment (GDP per capita) and the magnitude of the corresponding estimates through the analysis.  But as Gelman and Hill (2007) point out, the two-step analysis can cause problems when there are interactions between individual- and country-level predictors; and multilevel modeling, though similar to the two-step model, is a more general approach incorporating both levels at once (Gelman and Hill 2007).   As the purpose of this chapter is to test implications of the factor endowment model, the most relevant individual-level variable is a proxy for individual skill: years of education (education)51.  Besides the standard socioeconomic and structural variables such as income, age, gender, the self-placement on the left-right political scale, employment status, union membership, and marital status were added as control variables.  Because of the large number of missing data and different coding schemes across countries, it is difficult to arrive at a reliable and comparable measurement of individual income levels across 18 countries.  I thus included a subjective estimate of income levels of the respondent instead.  Respondents’ economic insecurity measures were also added to test the human capital hypothesis (HCH). I included two different types of economic insecurity measures – egocentric insecurity and sociotropic insecurity.  For an egocentric economic insecurity measure, I use survey responses to the following question: “How worried are you about your work situation?”; and for a sociotropic economic insecurity measure: “How worried are you about the economy?”. The responses are coded on a scale from 1 (not worried at all) to 4 (very worried).  Lastly, as non-economic independent variables, I looked at national pride and attitude toward income inequality.  For national pride, I focus on answers to the following question: “How proud are  51 Unfortunately I did not manage to use individual occupations as a proxy because it is complicated to classify occupations in a consistent manner across the countries.   165 you to be [Japanese]?”; and for attitude toward income inequality: “How much do you agree or disagree with the following statement: income should be made more equal”.   As the main independent variables on the country-level, I first included Real Gross Domestic Product (GDP) per Capita relative to the United States in the model as a proxy for a factor-endowment proportion of a country.  I included economic openness and growth rate of Real GDP Chain per capita from PWT as control variables at the country level.  I also created contextual variables – mean (and median) of education, mean of insecurity, mean of attitudes toward income inequality, and mean of national pride in each country – by averaging the individual level predictors above for each country.  Country mean of education is used as a proxy for a factor-endowment proportion, along with GDP per capita, because average (median) level of education reflects ratio of skilled to unskilled workers of a country. Country mean of insecurity is included to test the explanatory power of the Human Capital Hypothesis (HCH) relative to the H-O.  Economic insecurity has indeed been considered the most frequent and powerful rationale for protectionism (Irwin 2005a; Hays, Ehrlich, and Peinhardt 2005, 473).  If economic insecurity at the contextual level shapes trade attitudes – i.e., if people who reside in a country where (sociotropic) economic insecurity is substantially high (mostly LDCs) tend to be more protectionist than those from a country where economic insecurity is relatively low on average – it can be argued that what matters is not so much distributive implications of trade suggested by the H-O as sociotropic concerns about the country’s economy in general.  Table 5.1 lists the country-level (contextual) variables including the country means of education, insecurity, attitudes toward inequality, and national pride.   Table 5.1 Country-level independent variables Factor endowment proportions Other economic factors Economic Insecurity Non-economic factors Country GDP per capita 2001 (relative to the US) Education (Mean (Median)) Economic Openness (%) Equity Egocentric insecurity (Mean) Sociotropic insecurity (Mean) Attitude toward inequality (Mean) National pride (Mean) Japan              72.60147 3.212633 (3) 22.67941 4.5 2.036400  2.427366 3.240862 3.115349 Korea        48.80532 3.073267 (3) 64.19102 7.8 2.222564  2.653185 3.874104 3.130832 China              11.44182 2.886228 (3) 38.70758 13.2 1.997978  2.172589 2.932460 3.333333 Taiwan             57.40303 2.562874 (3) 110.3941 - 1.912475 2.420732 2.598746 2.848614 Singapore          89.54127 2.682903 (3) 342.1796 17.7 1.530488 1.655276 2.967314 3.375506 Malaysia           33.65490 2.577000 (3) 206.1021 11 1.701477 2.159465 3.092161 3.702106 Indonesia      11.21363 2.460697  (3) 54.81021 10.8 1.713666 2.440239 4.037111 3.367953 Thailand            18.55672 2.500000 (2) 139.2952 13.1 1.954774 2.555444 3.846540 3.876000 Philippines       10.61031 2.645000 (3) 98.22464 14.1 2.501006 2.641642 3.266800 3.849850 UK      68.66811 3.310865 (3) 51.88995 13.8 1.445918 1.734631 3.586626 3.211716 Ireland            81.80431 3.549505 (4) 154.4351 9.4 1.304392 1.596330 3.925000 3.629032 France             69.29531 3.534161 (4) 50.35385 9.1 1.553236 1.751037 4.344032 3.065979 Germany            73.69066 2.853085 (3) 61.34277 6.9 1.705348 1.902414 3.370887 2.719432 Sweden             71.12429 3.106893 (3) 84.69545 6.2 1.432544 1.692872 4.295132 3.063808 Italy                68.11229 3.110345 (3) 53.8563 11.6 1.706374 2.163265 4.455179 3.258000 Spain               61.67783 2.800200 (3) 52.61053 10.3 1.953299 2.227642 3.965625 3.182006 Portugal            48.88723 2.416000 (2) 60.5637 15 1.771717 2.160825 4.544814 3.479397 Greece              51.25273 2.563851 (3) 61.49029 10.2 2.272727 2.484970 4.421891 3.674651 Sources: GDP per capita (relative to the US), Economic openness are from PWT; Equity measure is derived from human development reports (United Nations Development Program, 2003).      167   To test the theories outlined above (the H-O and HCH), multilevel models that simultaneously estimate individual- and country-level effects are used.  The data are hierarchically organized with individuals nested within countries, and information at both the individual-level and the country-level is used to determine trade attitude.  I use the lmer ( ) function52 in R for the analyses.   For the baseline model, the full specification of the multilevel model is: Pr(y i = 1) = logit 1 (α ][ij ), for i = 1, …, n α j  ~ N (μα , σ 2 country ), for j = 1, …18.   5.4 Results and discussion  Table 5.2 lists five multilevel models.  The first is a null model that has the intercept only. Conducting multilevel regression analysis makes sense when there is a sufficient level of variation in the dependent variable at the country level.  According to the null model, individual-level variation in protectionist sentiment is 1.00 and country-level variation is 0.56 on the logit scale 53.  The intra-country correlation is therefore 0.36 on the logit scale54.  This indicates that protectionist sentiment differs not only within countries (across individuals) but also across countries; and about ±9%55 (on the probability scale) of the variation in protectionist sentiment can be explained by country-level characteristics.  The high  52 The name lmer stands for “linear mixed effects in R,” but the function actually works for generalized linear models as well.  The lmer function is part of R package Matrix. 53 Given that the standard deviation is 0.75, the variation is 0.75^2, which is 0.56. 54 The intra country correlation is: 0.56/(1.00+0.56) = 0.36. 55 The slope of the logistic curve – the derivative of the logistic function – is maximized at its center, which attains the value β/4.  Thus, 9% (=β/4) is the maximum difference in Pr(y = 1).   168 percentage of the variation at the country-level suggests that the multilevel character of protectionist sentiment should not be ignored.  Table 5.2 Multilevel logistic regression for protectionist sentiment: Factor endowment models Parameter Null Model Model 1 (Baseline) Model 2 Model 3 Model 4 Intercept -0.06 (0.18)  1.76 (0.44)*** -0.19 (1.51)  1.76 (0.45)  1.50 (0.68)* Individual level factors Demographic    Female   0.31 (0.05)***  0.31 (0.05)***  0.31 (0.05)***  0.31 (0.05)***    Education  -0.21 (0.06)*** -0.21 (0.06)*** -0.21 (0.06)*** -0.21 (0.06)***    Subjective Income (Rich)  -0.13 (0.15) -0.13 (0.15) -0.13 (0.15) -0.16 (0.15)    Age   0.04 (0.02)*  0.04 (0.02)*  0.04 (0.02)*  0.04 (0.01)*    Left-right ideology   0.05 (0.01)***  0.05 (0.01)***  0.05 (0.01)***  0.05 (0.02)***    Public sector employment   0.01 (0.06)  0.01 (0.06)  0.01 (0.06)  0.00 (0.06)    Unionmembership   0.13 (0.06)*  0.13 (0.06)*  0.13 (0.06)*  0.16 (0.06)**    Employment status   0.04 (0.05)  0.04 (0.05)  0.04 (0.05)  0.07 (0.05)    Single  -0.18 (0.07)* -0.18 (0.07)* -0.18 (0.07)* -0.18 (0.07)*  Country level factors    GDP per capita (relative to US)  -0.03 (0.01)***  -0.03 (0.01)*** -0.03 (0.01)***    Average level of education   0.08 (0.52)    Economic openness     0.00 (0.00)    Equity     0.02 (0.06)  Standard deviation    Level 1: Individuals  1.00 1.00 1.00  1.00  1.00    Level 2: Country intercept  0.75 1.00 0.78  1.01  1.06    Level 2: Education   0.20 0.20  0.20  0.21   Notes: Entries are results of multilevel logistic regression analyses, with a binary variable, protectionist sentiment, as dependent variable. Significant codes:   ***p < 0.001. **p < 0.01. *p < 0.05. .p < 0.1    170 The multilevel models presented in Table 5.2 assessed the effect of GDP per capita as a proxy for a factor endowment proportion of a country in a sample of eighteen countries across Europe and Asia.  Unlike the H-O prediction, the results show that the protectionist sentiment is higher among people living in capital-scarce countries.  The coefficient for GDP per capita (relative to the US) is -0.03, which tells us that 1 unit increase in GDP per capita relative to US (US = 100) corresponds to maximum 0.75%56 negative difference in the probability of being protectionist.  The negative effect of GDP per capita on protectionist sentiment is statistically significant even after economic openness and equity levels are controlled for (Model 3 and 4 in Table 5.2).  Basically, the models display little support for the H-O model.  The argument that people in labour abundant (and/or capital-scarce) countries want economic openness is empirically groundless; rather, the reverse is true: people in LDCs tend to be more protectionist than those in developed countries.  If for policy outcomes what matters is the median voter’s preferences (at least in democracies), as argued by Milner and Kubota (2005), O’Rourke and Taylor (2006), and Tavares (2006), the results suggest democratization likely lead to more protectionist economy, not to more open economy.  The results also question that education’s effects are representative of skill levels. The multilevel models with varying-intercept and varying-slope with the individual level predictor, education, display that educational attainment and protectionist sentiment are negatively associated in most countries (except for the three countries: Indonesia, Philippines, and Thailand), although the association is rather weak in some LDCs.  Even among the three countries that display a positive association between education and protectionist  56 0.0075 = 0.03/4   171 sentiment, the association is not strong (the coefficients are 0.07, 0.11, and 0.04, respectively). That is, the less educated the more protectionist in most countries.  Also, a country-mean (and median) of education as a proxy for a factor proportion of a given country (ratio of skilled to unskilled labour) has no effect on protectionist sentiment.  That is, there is little difference in protectionist sentiment between people living in countries abundant in high- skilled labour and people living in countries abundant in low-skilled labour.  This indicates either that educational attainment may not be representative of skill-levels, or that trade attitudes are not really guided by distributive concerns predicted by the H-O.   As the models presented in Table 5.2 suggest that individual economic gains and losses may not be a primary determinant of trade attitudes.  The multilevel analyses in Table 5.3 demonstrate that economic insecurities and attitudes toward income inequalities have statistically significant effects.  Being more worried about the country’s economy (sociotropic insecurity) by 1 unit on a scale 1 (not so worried) to 3 (very worried) is expected to increase maximum 8.25% (= 0.33/4) of the probability of being protectionist; and likewise, being more worried about the respondent’s work situation (egocentric insecurity) by 1 corresponds to maximum 2.75% (= 0.11/4) positive difference in the probability of being protectionist. Model 2 and 3 in Table 5.2 introduce the contextual variables, the country mean of the sociotropic insecurity and country mean of the egocentric insecurity respectively, as well as the individual level effects.  Model 2 demonstrates that the contextual variable matters too: people who reside in a country where concerns about the country’s economy are prevalent are more likely to be protectionist than those who do not.  When the country mean of sociotropic insecurity increases by 1, the probability of being protectionist increases by   172 maximum 16.75% (= 0.67/4).  Yet, the effect of the other contextual variable, the country mean of egocentric insecurity, is not statistically significant (Model 3).   Table 5.3 Multilevel logistic regression for protectionist sentiment: Inequality and insecurity Parameter Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Intercept -1.67 (0.39)*** -4.11 (0.90)*** -3.67 (0.89)*** -2.61 (0.26)*** -4.64 (1.00)*** -6.80 (1.28) Individual-level Factors Demographic    Female  0.26 (0.05)***  0.26 (0.05)***  0.26 (0.05)***  0.27 (0.05)***  0.26 (0.05)***  0.26 (0.05)***    Education -0.14 (0.02)***   -0.14 (0.02)***   -0.14 (0.02)***  0.73 (0.13)*** -0.14 (0.02)*** -0.14 (0.02)***    Subjective income (Rich) -0.04 (0.15)   -0.05 (0.15)   -0.05 (0.15)   -0.01 (0.15) -0.05 (0.15)   -0.05 (0.15)    Age  0.05 (0.02)**  0.05 (0.02)**    0.05 (0.02)**    0.05 (0.02)***  0.05 (0.02)**    0.05 (0.02)**    Left-right ideology  0.04 (0.01)***  0.04 (0.01)***   0.05 (0.01)***  0.05 (0.01)*  0.05 (0.01)***  0.04 (0.01)***    Public sector employment  0.03 (0.06)    0.03 (0.06)    0.03 (0.06)  0.03 (0.06)  0.03 (0.06)  0.03 (0.06)    Unionmembership  0.14 (0.06)*  0.13 (0.06)*    0.13 (0.06)*    0.14 (0.06)*  0.13 (0.06)*    0.14 (0.06)*    Employment status  0.00 (0.05)   0.00 (0.05)    0.00 (0.05)    0.00 (0.05)  0.00 (0.05)    0.00 (0.05)    Single -0.15 (0.07).   -0.15 (0.07)* -0.15 (0.07)* -0.16 (0.07)* -0.15 (0.07)* -0.14 (0.07). Economic Insecurity    Sociotropic insecurity  0.33 (0.04)***  0.34 (0.04)***  0.34 (0.04)***  0.33 (0.04)***  0.34 (0.04)***  0.34 (0.04)***    Egocentric insecurity  0.11 (0.04)**  0.11 (0.04)**  0.11 (0.04)**  0.12 (0.04)**  0.11 (0.04)**  0.11 (0.04)** Inequality    Attitude toward inequality  0.19 (0.02)***  0.19 (0.02)***  0.19 (0.02)***  0.19 (0.02)***  0.18 (0.02)***  0.19 (0.02)*** National Pride    National Pride  0.25 (0.03)***    0.25 (0.03)***    0.25 (0.03)***    0.25 (0.03)***  0.25 (0.03)***    0.25 (0.03)***  Country-level Factors    GDP per capita (relative to US) -0.02 (0.01)***    Mean of Sociotropic insecurity  0.67 (0.40).    Mean of Egocentric insecurity    0.56 (0.48)    Mean of Inequality      0.52 (0.26)*    Mean of National pride       1.25 (0.38)** Interactions    Mean of Sociotropic insecurity :    Education     0.27 (0.06)***  Standard deviation    Level 1: Individuals  1.00  1.00  1.00  1.00  1.00  1.00    Level 2: Country intercept  0.58  0.58  0.60  0.60  0.56  0.48    174  It is noteworthy that sociotropic insecurity matters more than egocentric insecurity. The relative importance of sociotropic insecurity to egocentric insecurity implies that trade attitudes are guided less by material self-interest than by perceptions of the country’s economy as a whole.  If that is a usual practice, protectionist sentiments can be inflated when individuals have little confidence in the working of domestic economic and political institutions, as I discussed in detail in the earlier chapters.  As shown in Table 5.1, the country means of sociotropic insecurity are a lot higher in LDCs than in developed countries: that is, unlike what the H-O model predicts, the electorates in LDCs more likely lean toward protectionism than those in developed countries, all else being equal.  In addition, the interaction term between the country-mean of sociotropic insecurity and educational attainment in Model 4 suggests that positive changes in a country-mean of sociotropic insecurity reduce education’s positive effect on protectionist sentiment.  In other words, while educational attainment has a negative association with protectionist sentiment in general, the negative association can be moderated in countries where concerns about the country’s economy prevail.  Although the empirical pattern that educational attainment and support for open trade are positively associated in developed countries but the association weakens or is even reversed in LDCs is often interpreted as support for the H-O model, the interaction term suggests that the association weakens in LDCs not so much because people base their judgments on the distributive implications of trade, as because people in LDCs are in general more likely to be concerned about the country’s economy.  Lastly, Model 6 and 7 suggest that relative income disparities (and perceived social injustice) and national pride can be an important motivation behind protectionist pressures.   175 Believing that society should reduce income inequalities has a positive effect on protectionist sentiment: people who are in favour of reduction of income disparities more likely lean toward protectionism than those who are not in favour.  Likewise, national pride is also found as a powerful determinant of protectionist sentiment.  Interestingly, both factors make a cross-country difference (as well as an individual level difference): as seen in Table 5.1, both attitudes toward inequality and national pride vary greatly among countries; and as Model 6 and 7 demonstrate, both contextual effects are statistically significant.  People living in a country where people in general believe that income should be more equally distributed are more likely to be protectionist.  It is interesting that the two developing countries, China and Taiwan, along with Singapore display the lowest mean score in the question about inequality (Table 5.1).  The strong positive relationship between being in favour of reduction of income disparities and protectionist sentiment seems to be significant in explaining why the three countries above all display surprisingly low levels of protectionist sentiment compared to the other LDCs.  Country-mean of national pride also varies greatly among countries, which is also significant in explaining the variation in protectionist sentiment.  5.5 Implications: Does democracy encourage trade openness?  Does democratization foster trade openness?  Setting aside the question whether public opinion affects trade policy, if the claim hinges on the H-O implication, the answer is no. The findings in this study demonstrate that it is not empirically valid that the electorates in capital scarce countries prefer greater trade openness.  If the assumption that the electorates want open trade were not valid as suggested, then the newly democratized governments would have little incentive to adopt trade liberalization policy; and if the electorates were   176 protectionist in general, then democratization, i.e., broadening the scope of population eligible to vote, would lead to more closed economy.   The findings in this study suggest that trade attitudes are not simply a function of who is personally benefited or hurt by trade in labour market, conditional on a factor endowment proportion of a given country.  Multilevel logistic regression analyses assessed the effect of both country-level and individual-level determinants in a sample of eighteen countries across Europe and Asia; and found that protectionist sentiment is significantly high among people living in low income economies, countries with the high level of economic insecurity, and countries where people in general believe income inequality should be reduced.  These findings imply that economic insecurity and perceived social injustice (income inequality) can be an important motivation behind protectionist pressures; and that protectionist forces more likely prevail in new democracies, which are often characterized by weak economic and political institutions, than in advanced democracies.     177 CONCLUSION  The welfare state is a bankruptcy law for workers.  In the same way bankruptcy laws encourage risk-taking by entrepreneurs, the welfare state encourages workers to be more open to change (and the resulting risks) in their attitudes – Ha-Joon Chang (2011)  Summary of the findings   Each of the five chapters in this dissertation tackled the following theoretical or empirical puzzle about protectionist sentiment: why is there a discrepancy in views on trade liberalization between economists and the public (Chapter 1); why are females more protectionist than males (Chapter 2); why does more spending on welfare bring about more support for openness in some countries but not in others (Chapter 3); what explains the recent protectionist backlash in Korea, an export-oriented economy where there seems to be a public consensus on the positive impact of trade on the national economy (Chapter 4); and does democratization lead to more economic openness as predicted by factor endowment models (Chapter 5)?   The first two chapters of the dissertation examined the effects of individual concerns about the alleged domestic social implications of trade policy on trade attitudes.  Chapter 1, “Distributional Judgment in Individual Preferences over Trade Liberalization,” provided an empirical test of how much individuals’ trade policy evaluations are affected by the policy’s distributional consequences, as a concomitant of income growth generated by trade liberalization.  By tackling the psychological implications of the trade-off between efficiency of income growth and equality of income distribution, this chapter found that concerns about inequality, poverty, and job prospects served as an effective countervailing force in shaping individual trade attitudes against the effect of aggregate benefits from free trade.   178 The findings suggest that if the enlarged economic pie through trade liberalization is unequally shared, the rise in average income is unlikely to help build public support for more openness.   Chapter 2, “Female Protectionism: The Consequence of Knowledge Gap in Economics?,” examined the sources of the female protectionism by testing the claim that exposure to economic ideas at university accounts for the gender gap in protectionist sentiment (Burgoon and Hiscox’s hypothesis).  The survey experiments found that the gender gap was not simply reducible to differences in education experiences: the gender gap remained strong even after controlling for levels of economic knowledge.  Drawing on the insights of public opinion scholars who have long suggested that public opinion is often guided by the sympathies of people feel toward victims of the policy, I tested and confirmed the hypothesis that the gender gap in protectionist sentiment may be generated primarily by the differing degrees of sympathy that men and women have for those groups implicated in the policy.   The last three chapters of the dissertation investigated institutional level factors shaping public attitudes toward trade policy.  Chapter 3, “The Compromise of Embedded Liberalism and Government Credibility,” provided evidence that public support for trade liberalization depends on the public’s trust in its government’s desire and ability to cushion trade-induced insecurity and inequality.  While trade liberalization is often accompanied by promises that losers from trade will be compensated, these promises are not binding. Presumably, people are more likely to support a greater degree of openness only when the government promises to cushion the adverse domestic effects of open markets are viewed as   179 credible.  The cross-national survey analyses found empirical evidence of a strong effect of government credibility on protectionist sentiment, suggesting that in LDCs that have never enjoyed the privilege of expansive social protection, and that are often characterized by lack of effective government institutions, the security-enhancing functions of government promises are not likely to work at the same level they do in countries with an effective government and a long welfare state tradition. .  Chapter 4, “Trust in Government and Protectionist Sentiment: Korean Public Opinion on the Korea-US Free Trade Agreement (KORUS-FTA),” demonstrated that the recent protectionist backlash in Korea can be also explained primarily by a lack of political trust in government’s desire and capacity to respond to public demands for effective policy implementation and equitable distribution of the economic pie.  Drawing on the political trust literature, which suggests that new policy initiatives that inevitably involve uncertainties and risks demand more trust from the public than those whose outcomes are relatively certain, I hypothesized that public opposition to the KORUS FTA in Korea in large part reflected the public condemnation of responsiveness and effectiveness of government institutions.  Content analysis of news media coverage confirmed that both the FTA and U.S. beef issues were framed in large part along storylines that raise public suspicion of government responsiveness and government effectiveness.  The majority of the criticisms of the trade initiatives in question were directed at the government (the President and negotiators), and the protectionist arguments evolved into arguments about a crisis of representative democracy.    180  In Chapter 5, “Does Democracy Encourage Trade Liberalization? Factor Endowments and Protectionist Sentiment in Less-Developed Countries,” I tested the Hecksher-Ohlin assumption that the electorates (the median voter) in LDCs are more likely to prefer lower levels of protection than those in developed countries.  Multilevel logistic regression analyses found that stronger protectionist sentiment prevailed in new democracies than in advanced democracies, counter to the predictions of the factor endowment models. Protectionist sentiment was significantly high among people living in capital-scarce countries, countries with high level of economic insecurity, and countries with a strong public demand for income redistribution.  The findings of these last three chapters of the dissertation suggest that given fragile economic and political institutions, protectionist forces more likely prevail in new democracies than in advanced democracies.  Implications for new democracies  The political economy literature, which looks at the impact of political regimes on trade liberalization, has largely concluded that democratization has a positive effect on trade liberalization.  While basing their arguments on the untested assumption that trade attitudes are largely a function of who personally benefits or is hurt by trade in the labour market, the existing arguments largely disregard the political dynamics surrounding a given country’s policy initiatives.  In fact, the emergence of democracy does not simply imply broadening the franchise to the wider population; it also implies changing patterns of political interactions between the government and the public.  That is, democratization affects trade policies not only by shifting the location of the median voter, but it also likely affects the policies by directly changing nature of political interactions between the governing elite and the public.   181 This dissertation in large part attempted to address political dilemmas faced by democratically elected politicians committed to trade liberalization in newly democratized countries, demonstrating that new democracies are not necessarily in a better position than their authoritarian counterparts to garner public support for trade liberalization.  The findings of this dissertation are in essence twofold: (1) the positive effect of income growth on support for trade can be significantly offset by concerns with the effect of trade liberalization on domestic social and economic arrangements; (2) public support for trade liberalization depends on the public’s trust in its government’s desire and ability to cushion trade-induced insecurity and inequality.  First, the importance of the communitarian critiques of trade liberalization in shaping individual trade attitudes not only suggests that trade-induced economic growth does not necessarily lead to public support for greater trade openness; but it also suggests that the prediction of the factor endowment models that the electorates (the median voter) in capital scarce countries prefer more openness is flawed. Second, declining political trust (since democratization) and fragile economic and political institutions in many new democracies also suggest that protectionist backlash likely gains considerable momentum in new democracies as shown in the case of Korea in Chapter 4. The literature explaining rapid economic growth under authoritarian regimes often emphasizes the autonomy of the state from society (Evans 1995).  Democratization, however, made it inconceivable for politicians and bureaucrats to insulate themselves from social pressure because it inevitably leads to the emergence of widespread political participation as a central objective in its own right.  The characteristics of the so-called “delegative democracy” (O'donnell 1994, 55), where the elected president attempts to rule through personalistic movements that often bypass intermediate political institutions, are   182 rampant in many new democracies, making the prospects for greater public support for government’s trade initiatives even grimmer in new democracies than in advanced democracies.  It has been noted that Koreans in recent years have been increasingly more preoccupied with the issue of fairness. “Justice: What’s the Right Thing to Do?” by Michael Sandel sold more than one million copies in Korean edition in 2010, a rarity for an academic work.  And Ha-Joon Chang’s recent critic of market fundamentalism, “23 Things They Don’t Tell You About Capitalism” has also been a bestseller since its Korean edition was released in 2011 (Sandel 2010; Chang 2011).  This phenomenon may not be very surprising given that the Korean government has increasingly committed itself to neoliberal economic and trade policies in recent years, and that these policies are often viewed as the major culprits behind the country’s increasing income inequality and job insecurity.  The findings of this dissertation have two broad policy implications for politicians in new democracies: (1) politicians committed to economic policies that inevitably involve a trade-off between efficiency of income growth and equality of income distribution may need to build a domestic social contract by which people contract with each other through the government to protect them against market-generated risks and uncertainty; and (2) at the same time they may also need to commit themselves to having fledgling institutions of representative democracy take root, thereby making credible promises to a broad range of citizens possible.    183 REFERENCES  Abdelal, Rawi and John G. Ruggie. 2009. The principles of embedded liberalism: Social legitimacy and global capitalism. In New perspectives on regulation, eds. D. A. Moss, J. A. Cisternino, The Tobin Project. Adorno, T. W., E. Frenkel-Brunswik, DJ Levinson, and R. N. Sanford. 1950. The authoritarian personality. New York. Harpers: 123-96. Adserà, Alícia, and Carles Boix. 2002. Trade, democracy, and the size of the public sector: The political underpinnings of openness. International Organization 56 (02): 229. Alesina, A., E. Glaeser, and B. Sacerdote. 2001. Why doesn't the United States have a European-style welfare state? Brookings Papers on Economic Activity 2001 (2): 187-254. Alt, J. E., J. Frieden, M. J. Gilligan, D. Rodrik, and R. Rogowski. 1996. The political economy of international trade. Comparative Political Studies 29 (6): 689. Amiel, Y., and F. A. Cowell. 1999. Thinking about inequality: Personal judgment and income distributions. Cambridge Univ Pr. Anderson, C. J., and J. Pontusson. 2007. Workers, worries and welfare states: Social protection and job insecurity in 15 OECD countries. European Journal of Political Research 46 (2): 211-35. Andreoni, J. 1990. Impure altruism and donations to public goods: A theory of warm-glow giving. The Economic Journal 100 (401): 464-77. Atkinson, A. B. 1970. On the measurement of inequality. Journal of Economic Theory 2 (3): 244- 63. Baker, A. 2003. Why is trade reform so popular in Latin America? A consumption-based theory of trade policy preferences. World Politics 55 (3): 423-55. ———. 2005. Who wants to globalize. Am.J.Polit.Sci. 49 : 924. Balistreri, E. J. 1997. The performance of the Heckscher-Ohlin-Vanek model in predicting endogenous policy forces at the individual level. Canadian Journal of Economics 30 (1): 1-17. Bannister, G. J., and K. Thugge. 2001. International trade and poverty alleviation. International Monetary Fund. Bartels, L. M. 1991. Constituency opinion and congressional policy making: The Reagan defense build up. The American Political Science Review 85 (2): 457-74.   184 Beaulieu, E. 2002. Factor or industry cleavages in trade policy? An empirical analysis of the Stolper–Samuelson theorem. Economics & Politics 14 (2): 99-131. Beaulieu, E., M. Benarroch, and J. Gaisford. 2004. Trade barriers and wage inequality in a north-south model with technology-driven intra-industry trade. Journal of Development Economics 75 (1): 113-36. Beaulieu, E., and M. Napier. 2008. Why are women more protectionist than men? . Becker, Gary Stanley. 1971. The economics of discrimination. Economics Research Studies of the Economics Research Center of the University of Chicago. 2d - ed. Chicago: University of Chicago Press. Bianco, William T. 1994. Trust : Representatives and constituents. Michigan Studies in Political Analysis. Ann Arbor: University of Michigan Press. Black, D. 1958. The theory of committees and elections. Blind, Pery K. 2006. Building trust in government in the twenty-first century: Review of literature and emerging issues. 7th Global Forum of Reinventing Government. Building Trust in Government 26-29 June 2007, Vienna, Austria. Boix, Carles. 1998. Political parties, growth and equality : Conservative and social democratic economic strategies in the world economy. Cambridge studies in comparative politics. Cambridge, U.K ; New York, N.Y.: Cambridge University Press. Bolton, G. E., and A. Ockenfels. 2000. ERC: A theory of equity, reciprocity, and competition. American Economic Review 90 (1): 166-93. Burgoon, B., and M. J. Hiscox. 2008. The gender divide over international trade: Why do men and women have different views about openness to the world economy? Cambridge, MA: Harvard University. ———. 2003. The mysterious case of female protectionism: Gender bias in attitudes toward international trade. Paper presented at Annual Meeting of the American Political Science Association, Philadelphia, PA, August 28th-31st. Burgoon, Brian. 2001. Globalization and welfare compensation: Disentangling the ties that bind. International Organization 55 (3) (Summer): 509-51. Burstein, P. 2003. The impact of public opinion on public policy: A review and an agenda. Political Research Quarterly 56 (1): 29. Cameron, M. A., and B. W. Tomlin. 2002. The making of NAFTA: How the deal was done. Cornell Univ Pr.   185 Cameron, David R. 1978. The expansion of the public economy: A comparative analysis. The American Political Science Review 72 (4) (Dec.): 1243-61. Caplan, Brayan. 2001. What makes people think like economists? Evidence on economic cognition from the "survey of Americans and economists on the economy" 44 (2): 395- 426. ———. 2002. Systematically biased beliefs about economics: Robust evidence of judgemental anomalies from the survey of Americans and economists on the economy. Economic Journal 112 (479): 433-58. ———. 2006. How do voters form positive economic beliefs? Evidence from the survey of Americans and economists on the economy. Public Choice 128 (3): 367-81. ———. 2007. The myth of the rational voter. Princeton University Press. Catterberg, Gabriela, and Alejandro Moreno. 2006. The individual bases of political trust: Trends in new and established democracies. International Journal of Public Opinion Research 18 (1) (March 1): 31-48. Chang, H. J. 2011. 23 things they don't tell you about capitalism. Bloomsbury Press. Citrin, Jack. 1974. Comment: The political relevance of trust in government. The American Political Science Review 68 (3) (Sep.): 973-88. Conover, Pamela Johnston. 1988. Feminists and the gender gap. The Journal of Politics 50 (04): 985-1010. ———. 1988. The role of social groups in political thinking. British Journal of Political Science 18 (01): 51. Conover, Pamela Johnston and Stanley Feldman. 1986. Emotional reactions to the economy: I'm mad as hell and I'm not going to take it anymore. American Journal of Political Science 30 (1) (Feb.): 50-78. Cook, E. A., and C. Wilcox. 1991. Feminism and the gender gap—A second look. The Journal of Politics 53 (04): 1111-22. Cox, Donald. 1987. Motives for private income transfers. Journal of Political Economy 95 (3): 508-46. Croson, R., and U. Gneezy. 2009. Gender differences in preferences. Journal of Economic Literature 47 (2): 448-74. De Witte, Hans. 1999. Job insecurity and psychological well-being: Review of the literature and exploration of some unresolved issues. European Journal of Work and Organizational Psychology 8 (2): 155.   186 Deininger, Klaus, and Lyn Squire. 1998. New ways of looking at old issues: Inequality and growth. Journal of Development Economics 57 (2): 259-87. Diamond, Larry Jay. 1999. Developing democracy : Toward consolidation. Baltimore: Johns Hopkins University Press. Dollar, D. 1992. Outward-oriented developing economies really do grow more rapidly: Evidence from 95 LDCs, 1976-1985. Economic Development and Cultural Change 40 (3): 523-44. Dollar, D., and A. Kraay. 2000. Growth is good for the poor. “World Bank. Development Research Group, Washington, DC Processed. ———. 2004. Trade, growth, and poverty. The Economic Journal 114 (493, Features) (Feb.): F22-49. Dominitz, Jeff, and Charles F. Manski. 1997. Perceptions of economic insecurity: Evidence from the survey of economic expectations. The Public Opinion Quarterly 61 (2) (Summer): 261-87. Downs, Anthony. 1957. An economic theory of democracy. New York: Harper. Easterlin, R. A. 2000. Where is economic growth taking us? Paper presented at Mount Holyoke Conference on the World Economy in the 21st century: Challenges and opportunities, Mout Holyoke, February ———. 1995. Will raising the incomes of all increase the happiness of all? Journal of Economic Behavior & Organization 27 (1): 35-47. Easton, David. 1965. A systems analysis of political life. New York: Wiley. Eichengreen, B., and D. Leblang. 2008. Democracy and globalization. Economics & Politics 20 (3): 289-334. Erikson, R. S., G. C. Wright, and J. P. McIver. 1993. Statehouse democracy: Public opinion and policy in the American states. Cambridge Univ Pr. Espinal, Rosario, Jonathan Hartlyn, and Jana Morgan Kelly. 2006. Performance still matters: Explaining trust in government in the Dominican Republic. Comparative Political Studies 39 (2) (March 1): 200-23. Evans, P. B. 1995. Embedded autonomy: States and industrial transformation. Princeton Univ Pr. Fehr, E., and K. M. Schmidt. 1999a. A theory of fairness, competition, and cooperation*. Quarterly Journal of Economics 114 (3): 817-68.   187 ———. 1999b. A theory of fairness, competition, and cooperation. Quarterly Journal of Economics 114 (3): 817-68. Fernandez, Raquel, and Dani Rodrik. 1991. Resistance to reform: Status quo bias in the presence of individual- specific uncertainty. The American Economic Review 81 (5) (Dec.): 1146-55. Fording, R. C. 1997. The conditional effect of violence as a political tactic: Mass insurgency, welfare generosity, and electoral context in the American states. American Journal of Political Science 41 (1): 1-29. Fox, John. 2003. Effect Displays in R for Generalised Linear Models, Journal of Statistical Software 15 (8): 1-9. Freeman, Richard B. 1995. The large welfare state as a system. The American Economic Review 85 (2, Papers and Proceedings of the Hundredth and Seventh Annual Meeting of the American Economic Association Washington, DC, January 6-8, 1995) (May): 16-21. Frey, B. S., and A. Stutzer. 2002. What can economists learn from happiness research? Journal of Economic Literature 40 (2): 402-35. ———. 1999. Measuring preferences by subjective well-being. Journal of Institutional and Theoretical Economics (JITE) 155 (4): 755-78. Fukuyama, Francis. 1995. Social capital and the global economy. Foreign Affairs 74 (5) (Sep. - Oct.): 89-103. Fuller, Dan, and Doris Geide-Stevenson. 2003. Consensus among economists: Revisited. The Journal of Economic Education 34 (4) (Fall): 369-87. Gamson, William A. 1968. Power and discontent. The Dorsey series in anthropology and sociology. Homewood, Ill.: Dorsey Press. Garrett, Geoffrey. 1998. Partisan politics in the global economy. Cambridge studies in comparative politics. Cambridge, UK ; New York: Cambridge University Press. Garrett, Geoffrey, and Peter Lange. 1995. Internationalization, institutions, and political change. International Organization 49 (4) (Autumn): 627-55. Garrett, Geoffrey, and Deborah Mitchell. 2001. Globalization, government spending and taxation in the OECD. European Journal of Political Research 39 (2) (03/01): 145-77. Gelman, A., and J. Hill. 2007. Data analysis using regression and multilevel/hierarchical models. Cambridge University Press Cambridge. Gidengil, E. 1995. Economic Man—Social woman? Comparative Political Studies 28 (3): 384.   188 Gilens, M. 1988. Gender and support for Reagan: A comprehensive model of presidential approval. American Journal of Political Science 32 (1): 19-49. Gilligan, C. 1982. In a different voice: Psychological theory and women's development. Harvard Univ Pr. Gilligan, M. J. 1997. Empowering exporters: Reciprocity, delegation, and collective action in American trade policy. University of Michigan Press. Grossman, G. M., and E. Helpman. 1996. Electoral competition and special interest politics. The Review of Economic Studies 63 (2): 265. ———. 1992. Innovation and growth in the global economy. MIT press. Hagopian, F. 1990. “Democracy by undemocratic means”? Comparative Political Studies 23 (2): 147. Hainmueller, J., and M. J. Hiscox. 2007. Educated preferences: Explaining attitudes toward immigration in europe. International Organization 61 (02): 399-442. ———. 2006. Learning to love globalization: Education and individual attitudes toward international trade. International Organization 60 (02): 469-98. Hall, Peter A., and David W. Soskice. 2001. Varieties of capitalism : The institutional foundations of comparative advantage. Oxford England ; New York: Oxford University Press. Harrison, A. 2007. Globalization and poverty: An introduction. In Globalization and poverty. University of Chicago Press. Harsanyi, J. C. 1953. Cardinal utility in welfare economics and in the theory of risk-taking. The Journal of Political Economy 61 (5): 434. Hart, Roderick P., and Daron Shaw. 2001. Communication in U.S. elections : New agendas. Lanham, Md.: Rowman & Littlefield. Hausman, D. M., and M. S. McPherson. 2006. Economic analysis, moral philosophy, and public policy. Cambridge Univ Pr. Hays, S. P., M. Esler, and C. E. Hays. 1996. Environmental commitment among the states: Integrating alternative approaches to state environmental policy. Publius: The Journal of Federalism 26 (2): 41. Hays, Jude C., Sean D. Ehrlich, and Clint Peinhardt. 2005. Government spending and public support for trade in the OECD: An empirical test of the embedded liberalism thesis. International Organization 59 (02): 473. Hejazi, W., and A. E. Safarian. 1999. Trade, foreign direct investment, and R&D spillovers. Journal of International Business Studies: 491-511.   189 Herrmann, R. K., P. E. Tetlock, and M. N. Diascro. 2001. How Americans think about trade: Reconciling conflicts among money, power, and principles. International Studies Quarterly 45 (2): 191-218. Hetherington, Marc J. 2005. Why trust matters : Declining political trust and the demise of American liberalism. Princeton, NJ: Princeton University Press. ———. 1998. The political relevance of political trust. The American Political Science Review 92 (4) (Dec.): 791-808. Hetherington, Marc J., and Suzanne Globetti. 2002a. Political trust and racial policy preferences. American Journal of Political Science 46 (2) (Apr.): 253-75. ———. 2002b. Political trust and racial policy preferences. American Journal of Political Science 46 (2) (Apr.): 253-75. Hiscox, M. J. 2006. Through a glass and darkly: Attitudes toward international trade and the curious effects of issue framing. International Organization 60 (03): 755-80. ———. 2001. Class versus industry cleavages: Inter-industry factor mobility and the politics of trade. International Organization 55 (01): 1-46. Irwin, Douglas A. 2005. Free trade under fire. Princeton paperbacks. 2nd ed. Princeton, NJ: Princeton University Press. Jacobs, Alan. 2005. A matter of trust: Cognition, institutions, and the sources of credible commitment. Paper presented at the American Political Science Association annual meeting, Washington, DC. Jacobs, Lawrence R., and Robert Y. Shapiro. 1994. Studying substantive democracy. PS: Political Science and Politics 27 (1) (Mar.): 9-17. ———. 1995. Don't blame the public for failed health care reform. Journal of Health Politics, Policy and Law 20 (2): 411. ———. 2000. Politicians don't pander : Political manipulation and the loss of democratic responsiveness. Studies in communication, media, and public opinion. Chicago, IL.: University of Chicago Press. Johnston, R., K. Banting, W. Kymlicka, and S. Soroka. 2010. National identity and support for the welfare state. Canadian Journal of Political Science/Revue Canadienne De Science Politique 43 (02): 349-77. Johnston, Richard. 1992. Letting the people decide : Dynamics of a Canadian election. Stanford, Calif.: Stanford University Press.   190 Jon A. Krosnick. 1991. The stability of political preferences: Comparisons of symbolic and nonsymbolic attitudes. American Journal of Political Science 35 (3) (Aug.): 547-76. Jones, Bryan D. 2003. Bounded rationality and political science: Lessons from public administration and public policy. Journal of Public Administration Research and Theory 13 (4) (October 1): 395-412. Jurenas, R., and M. E. Manyin. 2010. US-South Korea beef dispute: Issues and status. Kapstein, E. B. 2007. Economic justice in an unfair world: Toward a level playing field. Princeton Univ Pr. Kapstein, E. B., and N. Converse. 2008. The fate of young democracies. Cambridge University Press. Katzenstein, Peter J. 1985. Small states in world markets : Industrial policy in Europe. Cornell studies in political economy. Ithaca, N.Y.: Cornell University Press. Kaufman, Robert R., and Alex Segura-Ubiergo. 2001. Globalization, domestic politics, and social spending in Latin America: A time-series cross-section analysis, 1973-97. World Politics 53 (4) (Jul.): 553-87. Keefer, P. 2007. Clientelism, credibility, and the policy choices of young democracies. American Journal of Political Science 51 (4): 804-21. Keefer, P., and R. Vlaicu. 2008. Democracy, credibility, and clientelism. Journal of Law, Economics, and Organization 24 (2): 371. Keele, Luke. 2005. The authorities really do matter: Party control and trust in government. The Journal of Politics 67 (3) (Aug.): 873-86. Kevin O'Rourke. 2007. Democracy and protectionism. IIIS, . Kinder, D. R. 1986. The continuing American dilemma: White resistance to racial change 40 years after myrdal. Journal of Social Issues 42 (2): 151-71. Kinder, D. R., and D. R. Kiewiet. 1981. Sociotropic politics. British Journal of Political Science 11 (2): 129-61. Kinder, D. R., and L. M. Sanders. 1996. Divided by color: Racial politics and democratic ideals. University of Chicago Press. ———. 1987. Pluralistic foundations of American opinion on race. Paper presented at annual meeting of the American Political Science Association, Chicago, .   191 Kinder, Donald R. 1998. Opinion and action in the realm of politics. In The handbook of social psychology., eds. Daniel T. Gilbert, Susan T. Fiske and Gardner Lindzey. 4th ed. ed. Vol. Vols. 1 and 2 (4th ed.). New York, NY, US: McGraw-Hill. Kinder, D. R., and D. R. Kiewiet. 1979. Economic discontent and political behavior: The role of personal grievances and collective economic judgments in congressional voting. Am.J.Polit.Sci. 23 : 495. Kittel, Bernhard, and Hannes Winner. 2005. How reliable is pooled analysis in political economy? the globalization-welfare state nexus revisited. European Journal of Political Research 44 (2): 269-93. Kluegel, J. R., and E. R. Smith. 1986. Beliefs about inequality: Americans' views of what is and what ought to be. Aldine de Gruyter. ———. 1982. Whites' beliefs about blacks' opportunity. American Sociological Review 47 (4): 518-32. Kollman, K. 1998. Outside lobbying: Public opinion and interest group strategies. Princeton University Press. Kono, D. Y. 2008. Democracy and trade discrimination. The Journal of Politics 70 (04): 942-55. ———. 2008. Does public opinion affect trade policy? Business and Politics 10 (2): 1224. Krueger, Anne. 1989. Asymmetries in policy between exportables and import-competing goods. Working Paper No. 2904 (March). Lau, R. R., and D. O. Sears. 1981. Cognitive links between economic grievances and political responses. Political Behavior 3 (4): 279-302. Lee, S. 2007. Democratic transition and the consolidation of democracy in south korea. Taiwan Journal of Democracy 3 (1): 99-125. Lewis-Beck, M. S. 1990. Economics and elections: The major western democracies. Univ of Michigan Pr. Lippmann, Walter. 1955. Essays in the public philosophy. --. 1 ] -- ed. Boston: Little, Brown. Lundberg, M., and L. Squire. 1999. Growth and inequality: Extracting the lessons for policymakers. Unpublished Paper, the World Bank. MacKuen, M. B., R. S. Erikson, and J. A. Stimson. 1992. Peasants or bankers? The American electorate and the U.S. economy. Am.Polit.Sci.Rev. 86 : 597. Mansfield, E. D., and D. C. Mutz. 2009. Support for free trade: Self-interest, sociotropic politics, and out-group anxiety. International Organization 63 (03): 425-57.   192 Mara, Gerald M. 2001. Thucydides and plato on democracy and trust. The Journal of Politics 63 (3) (Aug.): 820-45. March, James G., and Johan P. Olsen. 1989. Rediscovering institutions : The organizational basis of politics. New York: Free Press. Marcus, George E., W. Russell Neuman, and Michael MacKuen. 2000. Affective intelligence and political judgment. Chicago: University of Chicago Press. Mares, Isabela. 2004. Economic insecurity and social policy expansion: Evidence from interwar Europe. International Organization 58 (4) (Autumn): 745-74. Mayda, A. M., and D. Rodrick. 2005. Why are some people (and countries) more protectionist than others. Eur.Econ.Rev. 49 : 1393. Mayda, Anna Maria, and Dani Rodrik. 2005. Why are some people (and countries) more protectionist than others? European Economic Review 49 (6) (8): 1393-430. Mayer, W. 1984. Endogenous tariff formation. The American Economic Review: 970-85. Mayer, W., and J. Li. 1994. Interest groups, electoral competition, and probablistic voting for trade policies. Economics & Politics 6 (1): 59-77. McGraw, Kathleen M., Milton Lodge, and Jeffrey M. Jones. 2002. The pandering politicians of suspicious minds. The Journal of Politics 64 (2) (May): 362-83. Meltzer, Allan H., and Scott F. Richard. 1983. Tests of a rational theory of the size of government. Public Choice 41 (3): 403-18. Merton, R. K. 1957. Social theory and social structure (rev. Miller, Arthur H. 1974. Political issues and trust in government: 1964-1970. The American Political Science Review 68 (3) (Sep.): 951-72. Milner, H. V., and B. Mukherjee. 2009. Democratization and economic globalization. Annual Review of Political Science 12 : 163-81. Milner, Helen V., and Keiko Kubota. 2005. Why the move to free trade? Democracy and trade policy in the developing countries. International Organization 59 (1) (Winter): 107-43. Mitchell, G., P. E. Tetlock, B. A. Mellers, and L. D. Ordonez. 1993. Judgments of social justice: Compromises between equality and efficiency. Journal of Personality and Social Psychology 65 (4): 629. Monroe, A. D. 1998. Public opinion and public policy, 1980-1993. Public Opinion Quarterly 62 (1): 6.   193 Mughan, Anthony, and Dean Lacy. 2002. Economic performance, job insecurity and electoral choice. British Journal of Political Science 32 (3) (Jul.): 513-33. Munck, G. L., and C. S. Leff. 1997. Modes of transition and democratization: South America and Eastern Europe in comparative perspective. Comparative Politics 29 (3): 343-62. Mutz, D. C. 1992. Impersonal influence: Effects of representations of public opinion on political attitudes. Political Behavior 14 (2): 89-122. Mutz, D. C. 1998. Impersonal influence: How perceptions of mass collectives affect political attitudes. Cambridge studies in political psychology and public opinion. Cambridge ; New York: Cambridge University Press. Mutz, D. C., and J. J. Mondak. 1997. Dimensions of sociotropic behavior: Group-based judgements of fairness and well-being. American Journal of Political Science 41 (1): 284-308. Nelson, Thomas E., and Donald R. Kinder. 1996. Issue frames and group-centrism in American public opinion. The Journal of Politics 58 (4) (Nov.): 1055-78. North, D. C., and B. R. Weingast. 1989. Constitutions and commitment: The evolution of institutions governing public choice in seventeenth century england. J.Econ.Hist. 49(4) : 803. Douglass C. North, "Institutions and Credible Commitment" 1994), http://hdl.handle.net/10535/3711. Norton, M. I., and D. Ariely. 2011. Building a better America—One wealth quintile at a time. Perspectives on Psychological Science 6 (1): 9. O'donnell, G. 1994. Delegative democracy. Journal of Democracy 5 : 55-. Offe, C. 1999. "How can we trust our fellow citizens?" in Democracy and Trust, edited by M. E. Warren. Cambridge, England: Cambridge University Press. Olson, M. 1971. The logic of collective action: Public goods and the theory of groups. Harvard Univ Pr. O'Rourke, K. H., and A. M. Taylor. 2006. Democracy and protectionism. NBER Working Paper. O'Rourke, Kevin H., Richard Sinnott, J. David Richardson, and Dani Rodrik. 2001. The determinants of individual trade policy preferences: International survey evidence [with comments and discussion]. Brookings Trade Forum: 157-206. Page, Benjamin. I., and Robert. Y. Shapiro. 1983. Effects of public opinion on policy. The American Political Science Review 77 (1): 175-90.   194 ———. 1992. The rational public : Fifty years of trends in Americans' policy preferences. American politics and political economy series. Chicago: University of Chicago Press. Park, C. M. 2004. Support for democracy in Korea: Its trends and determinants. Paper presented at Conference on “The State of Democracy in Asia.” Bangkok, Thailand, . Pierson, Paul. 1994. Dismantling the welfare state? : Reagan, Thatcher, and the politics of retrenchment. Cambridge studies in comparative politics. Cambridge, England; New York: Cambridge University Press. Polanyi, K. 2001. The great transformation: The political and economic origins of our time. Beacon Pr. Putnam, R. 1988a. Diplomacy and domestic politics: The logic of two-level games. Int.Org. 48 : 427. Putnam, Robert D. 1988b. Diplomacy and domestic politics: The logic of two-level games. International Organization 42 (03): 427. Rabin, M. 1998. Psychology and economics. Journal of Economic Literature 36 (1): 11-46. Rawls, J. 1999. A theory of justice. Belknap Press. Rodrik, D. 1997. Has globalization gone too far? Peterson Institute. ———. 1997. Has globalization gone too far?. Washington, D.C.: Institute of International Economics. ———. 1998. Why do more open economies have bigger governments? The Journal of Political Economy 106 (5) (Oct.): 997-1032. Rogowski, R. 1989. Commerce and coalitions. Princeton University Press. Romer, P. M. 1989. What determines the rate of growth and technological change? Country Economics Dept., World Bank, . Rudra, Nita. 2002. Globalization and the decline of the welfare state in less-developed countries. International Organization 56 (2) (Spring): 411-45. Rudra, Nita, and Stephan Haggard. 2005. Globalization, democracy, and effective welfare spending in the developing world. Comparative Political Studies 38 (9) (November 1): 1015- 49. Ruggie, John Gerard. 1996. Globalization and the embedded liberalism compromise: The end of an era? Max Planck Institute for the Study of Societies. ———. 1983. Review: Continuity and transformation in the world polity: Toward a neorealist synthesis. World Politics 35 (2) (Jan.): 261-85.   195 Sachs, J. D., and A. Warner. 1995. Economic convergence and economic policies. Samuelson, P. A. 1950. Evaluation of real national income. Oxford Economic Papers 2 (1): 1-29. Sandel, M. J. 2010. Justice: What's the right thing to do? Farrar Straus & Giroux. Sapiro, Virginia. 2003. Theorizing gender in political psychology research. oxford handbook of political psychology. In Oxford handbook of political psychology., eds. David O. (Ed) Sears, Leonie (Ed) Huddy and Robert (Ed) Jervis. x ed. Vol. x, 601-634. New York, NY, US: Oxford University Press. Scheve, Kenneth F., and Matthew J. Slaughter. 2001. Globalization and the perceptions of american workers. Washington, D.C.: Institute for International Economics. ———. 2001. What determines individual trade-policy preferences? Journal of International Economics 54 (2) (8): 267-92. ———. 2004. Economic insecurity and the globalization of production. American Journal of Political Science 48 (4): 662-74. Schlozman, K. L., and S. Verba. 1979. Injury to insult: Unemployment, class, and political response. Harvard Univ Pr. Schonhardt-Bailey, C. 2006. From the corn laws to free trade: Interests, ideas, and institutions in historical perspective. The MIT Press. Sears, D. O., and C. L. Funk. 1990. The limited effect of economic self-interest on the political attitudes of the mass public. Journal of Behavioral Economics 19 (3): 247-71. Sears, D. O., and D. R. Kinder. 1985. Whites' opposition to busing: On conceptualizing and operationalizing group conflict. Journal of Personality and Social Psychology 48 (5): 1141-7. Sears, D. O., and R. R. Lau. 1983. Inducing apparently self-interested political preferences. American Journal of Political Science 27 (2): 223-52. Sears, D. O., R. R. Lau, T. R. Tyler, and H. M. Allen Jr. 1980. Self-interest vs. symbolic politics in policy attitudes and presidential voting. The American Political Science Review 74 (3): 670-84. Sen, A. K. 1977. Rational Fools: a critique of the behavioural foundations of economic theory. Philosophy and Public Affairs 6 (4): 317-344. Shapiro, Robert Y., and Harpreet Mahajan. 1986. Gender differences in policy preferences: a summary of trends from the 1960s to the 1980s. Public Opinion Quarterly 50 (1): 42-61. Shin, Gi-Wook. 2010. One alliance, two lenses : U.S.-Korea relations in a new era. Shorenstein asia- pacific research center series. Stanford, Calif.: Stanford University Press.   196 Simon, Herbert A. 1991. Bounded rationality and organizational learning. Organization Science 2 (1, Special Issue: Organizational Learning: Papers in Honor of (and by) James G. March): 125-34. Skocpol, Theda. 1992. Protecting soldiers and mothers: The political origins of social policy in the united states. Cambridge, Mass.: Belknap Press of Harvard University Press. Sniderman, P. M., R. A. Brody, and P. E. Tetlock. 1991. Reasoning and choice: Explorations in social psychology. Steenbergen, M. R., and B. S. Jones. 2002. Modeling multilevel data structures. American Journal of Political Science 46 (1): 218-37. Stiglitz, J. E., A. Sen, and J. P. Fitoussi. 2009. Report by the commission on the measurement of economic performance and social progress. Commission on the Measurement of Economic Performance and Social Progress, Paris. Stimson, J. A., M. B. MacKuen, and R. S. Erikson. 1995. Dynamic representation. The American Political Science Review 89 (3): 543-65. Stokes, B., P. Choate, and Council on Foreign Relations. 2001. Democratizing US trade policy. Council on Foreign Relations. Swank, Duane. 2001. Mobile capital, democratic institutions, and the public economy in advanced industrial societies. Journal of Comparative Policy Analysis 3 (2) (08/01): 133-62. Tavares, J. 2008. Trade, factor proportions, and political rights. The Review of Economics and Statistics 90 (1): 163-8. Walstad, W. 2002. The effects of teacher programs on student economic understanding and market attitudes in transition economies. In Reforming economics and economics teaching in the transition economies., eds. Michael Watts, William B. Walstad, 63–96. Warren, Mark. 1999. Democracy and trust. Cambridge, UK ; New York: Cambridge University Press. Weingast, B. R. 1995. The economic role of political institutions: Market-preserving federalism and economic development. Journal of Law, Economics, and Organization 11 (1): 1. Welch, S., and J. Hibbing. 1992. Financial conditions, gender, and voting in American national elections. The Journal of Politics 54 (01): 197-213. Williamson, O. E. 1985. The economic institutions of capitalism: Firms. Markets, Relational Contracting, New York.   197 Williamson, O. E. 1993. Calculativeness, trust, and economic organization. J.Law Econ. 34 : 453. Winters, L. Alan, Neil McCulloch, and Andrew McKay. 2004. Trade liberalization and poverty: The evidence so far. Journal of Economic Literature 42 (1) (Mar.): pp. 72-115. Wolpert, R. M., and J. G. Gimpel. 1998. Self-interest, symbolic politics, and public attitudes toward gun control. Political Behavior 20 (3): 241-62. Yang, C. C. 1995. Endogenous tariff formation under representative democracy: A probabilistic voting model. The American Economic Review 85 (4): 956-63. Zaller, John. 1992. The nature and origins of mass opinion. Cambridge England; New York, NY, USA: Cambridge University Press. Zhu, Yunhan. 2008. How East Asians view democracy. New York: Columbia University Press. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0072187/manifest

Comment

Related Items