1

Explaining Mass Support for Agricultural Protectionism

Evidence from a Survey Experiment During the Global Recession

Version 5.0: January 2010

Megumi Naoi Ikuo Kume

Assistant Professor Professor

Department of Political Science Faculty of Political Science & Economics

University of California, San Diego Waseda University, Tokyo

megumi.naoi@gmail.com kume@waseda.jp

Abstract: Why do citizens in advanced industrialized countries bear the high price of

agricultural products? Conventional wisdom suggests that agricultural interests secure

government protection because producers are concentrated and better politically organized

than diffused consumers. Due to its focus on producer capacity for collective action,

however, the literature fails to account for the high levels of mass support for agricultural

protectionism in advanced industrialized nations. This paper presents new evidence from a

survey experiment in Japan conducted during the current global recession (December 2008)

that accounts for this puzzle. Using randomly assigned visual stimuli, the experiment

activates respondents’ identification with either producer or consumer interests and

proceeds to ask attitudinal questions regarding food imports. The results suggest that

consumer-priming has no reductive or additive effects on the respondents’ support for

liberalizing food imports. Surprisingly, the producer-priming increases respondents’

opposition to food import, particularly among those who fear future job insecurity. We

further disentangle the puzzling finding that consumers think like producers for the issue of

food import along two mechanisms: “sympathy” for farmers and “projection” of their own

job insecurities. The results lend strong support to the projection hypothesis.

*We thank JSPA Grants-in-Aid for Scientific Research (A: Globalization and Domestic Politics, 2008-2012)

for financial support and Kiichiro Arai, Rieko Kage, Arata Kuno, Yoshitaka Nishizawa, Masahiko

Tatebayashi, Masaru Kohno, Kazunobu Hayakawa, Kengo Soga, Hiroki Mori, Yoshiko Kojo for their

comments and suggestions through the formulation and execution of this project and Kazumi Shimizu, Skip

Lupia, and Motoki Watanabe for being generous with their expertise on survey experiments. Participants at

GLOPE symposium at Waseda University, Kobe University, European University Institute, and IPSA

meeting in Santiago, Chile, UCLA conference on Post-bubble Japanese political economy, and IPES at Texas

A&M provided excellent comments. Naoi thanks the SSRC/Abe fellowship for financial support and Waseda

University for hosting her sabbatical when this research was conducted. Kenji Hall and Kim Chang-Ran

provided several photographs that were used as visual stimuli. Celeste Raymond Beesley provided

incomparable research assistance.

2

Introduction

Why do citizens in advanced industrialized nations bear the high price of

agricultural products? Despite a massive decline in the number of agricultural workers

and active farmland over time, agricultural protectionism is well and alive among

developed economies.1

OECD estimates that, on average, consumers in advanced

industrialized countries spend 10% of their annual consumption on agricultural products to

support farmers. The level of protection in Japan, however, stands out from the crowd. In

2008, Japanese consumers spent 41% of their annual expenditures on agricultural products

to support farmers (OECD 2009).2 [Figure 1 around here]

Not only are the actual levels of agricultural protection in Japan substantially

higher than other OECD countries, but a series of public opinion surveys, which directly

gets at individual preferences for regulating food imports, suggests that more than half of

Japanese citizens seem to be willing to bear this cost. A nationally-representative public

opinion survey we conducted in February 2009, in the midst of world-wide recession, asked

respondents to take a position on a trade-off between agricultural protectionism and

manufacturing export: “On import liberalization of agricultural products, there are

following two opinions, A and B. Which is closer to yours?”

1 Davis 2003, 2004; Gawande and Hoekman 2006; Park and Jensen 2007; Thies and Porche 2008.

2 OECD also calculates Producer Support Estimates (PSE), which are equal to the annual monetary

value of gross transfers from consumers and taxpayers to support agricultural producers in each

country. In 2006-2008, PSEs per gross farm receipts for the U.S., E.U., and Japan are 10%, 27%

and 49% respectively.

3

A: We should not accept import liberalization of agricultural products in order to

protect Japanese agriculture.

B: We should accept import liberalization of agricultural products in order to

maintain Japanese manufacturing export.3

Over the half of the respondents (55.6%) chose A, while 37.8% chose B, despite

the fact that farmers constitute only 3.9% of our sample and 6% of the total labor force4

in

Japan.5

Moreover, four waves of public opinion surveys on food import show that

Japanese citizens have become more protectionist over time. The Cabinet Office has

repeatedly asked a nationally-representative sample of respondents whether they would

support more food import “if foreign products are cheaper” or whether they support more

domestic production “even if it means higher prices.” In 1987, 19.9% supported more

food import if foreign products are cheaper and 71.2% support more domestic production

“even if it means higher prices.” The ratio of support for cheaper import dropped steadily

since then and in 2008, only 3.1% support cheaper import and 51.5% support domestic

production “even if it means higher prices.” [Figure 2 around here] This fact is harder to

3 Although this question poses a potential double-barreled problem, it has advantage as it presents a

trade-off between manufacturing export and agricultural protectionism. 4 The 6% figure includes part-time farmers (kengyo nouka) who are defined as agricultural workers

whose primary source of income is not farming, but spend at least 60-days per year on farming. 5 This survey was conducted by Waseda University Global COE Program. Farmers constitute 3.9%

of the total respondents and among farmers, 90% support A. The same question was asked in 1993

just before the general election by the Japan Election Study II. In this survey, 39.2% support A,

while 50.2% support B. The distribution is reversed, but still it is notable that substantial number

of Japanese citizens supported the agricultural protection.

4

grasp in light of urbanization and industrialization,6 the recession of 1990s, liberalization

through Uruguay Round of 1994,7 and the world-wide financial crisis today. Consumers,

especially lower-income consumers, should demand cheaper agricultural imports.

High public support for agricultural protectionism during the global recession,

moreover, is not confined to the Japanese case. A public opinion survey conducted in the

U.S. during March 2009, in the midst of the recession, suggests that 48% of respondents

think “it is a bad thing that Obama administration will cut down agricultural subsidies,”

while 40% of them think “it is a good thing”.8 The Eurobarometer conducted in the fall of

2007 shows that 50% of European respondents support the status-quo level of tariffs and

quota protection for agricultural commodities, while 37% oppose it.9

Two common explanations for agricultural protectionism are unhelpful to make

sense of this puzzle. The first focuses on the collective action capacity of interest groups:

producers (i.e., farmers) are concentrated and better politically organized than diffused and

unorganized consumers. The second focuses on political mobilization by elites: the

Liberal Democratic Party of Japan has been exchanging trade protection and subsidies for

rural and agricultural votes.10

Due to its focus on collective action capacity and organized

votes, however, the literature simply makes assumptions about consumers’ preferences for

7 Davis 2004.

8 Pew Research Center 2009.

9 Eurobarometer 2008.

10 Tachibana 1984; Kabashima 1984; Mulgan 2000; Davis 2004; Scheiner 2006.

5

free trade: consumers want free trade, but cannot act on it due to the collective action

problem. The two conventional accounts do not help us understand why the public seems to

be willing to bear high priced agricultural products to support farmers.

This paper presents new evidence from a survey experiment in Japan conducted

in December 2008 that challenges the assumption of free-trading consumers. The

experiment randomly assigns visual stimuli to activate respondents’ identification with

either producer or consumer interests and proceeds to ask attitudinal questions regarding

trade and food imports. The results suggest that activation of a consumer perspective is

associated with higher support for free trade on the issue of general trade, but that

consumer-priming has no systematic effect on the issue of food import. Surprisingly, the

producer-priming increases respondents’ opposition to food import particularly among

those who fear their own future job insecurity. We further propose two mechanisms to

explain why thinking about jobs and production activities makes respondents more

supportive of agricultural protectionism: “sympathy” for farmers and “projection” of their

own job insecurities onto farmers. The results lend strong support to the projection

hypothesis: those who fear future job insecurity and loss of income are the ones who

become more supportive of agricultural protectionism with the activation of a producer

perspective. This is paradoxical as workers with high job insecurity should be the prime

beneficiaries of cheaper food import.

6

Our results suggest that the mass public supports agricultural protectionism not

because they are concerned about quality and safety issues in food import from a consumer

perspective. Rather, one critical source of this support comes from the public’s projection

of their own job insecurity onto a symbolic declining sector, agriculture. The results help us

solve the paradox of persisting mass support for agricultural protectionism in the midst of

the world-wide recession. They also force scholars to move beyond the dichotomous

conceptualization of producer and consumer interests in the political economy literature

and pay due attention to why consumers often align with producers to support

protectionism even if it burdens them financially. Our finding that there exists an emerging

alliance of producers and protectionist consumers pushes forward the long-standing

research on coalition politics in the global economy which are mostly about class or

sectoral alliance.11

Beyond the literature on trade policy preferences and coalitions, the results have

broader implications for the study of inequality and redistribution in the global economy,

preference formation of mass publics on economic policy, and understanding how people

form “self-interest” and inferences about others’ needs in a social context.

The Puzzle: High Support for Agricultural Protectionism among the Public

Despite the fact that Japanese trade policy is often depicted as protectionist,

Japan’s average applied tariff rates have been one of the lowest among developed

11

Rogowski 1989; Hiscox 2002.

7

economies since the GATT Kennedy Round.12

Agriculture, however, is a very different

story: it still remains highly protected as shown in the ranking of Consumer Support

Estimates among OECD countries (see, Graph 1).

Not just the levels of protectionism, but the forms of compensating farmers in

Japan differ from the United States and E.U. Instead of compensating income loss to

farmers from liberalization with lump-sum tax transfers, the Japanese government

extensively uses price support and tariff and non-tariff barriers. For instance, rice, sugar,

and butter have the extremely high tariff rates of 778%, 270% and 330% respectively

(WTO 2008). These forms of protection directly tax consumers who have the opportunity

to feel this financial burden on a daily-basis. Yet, consumers seem to be willing to support

agricultural protectionism. This is particularly puzzling because when it comes to general

trade issues, Japanese citizens show the highest level of support (77%) for free trade among

twelve developed and developing economies.13

A nationally-representative public opinion survey we conducted in February 2009

indeed shows continuing mass support for agricultural protectionism. The survey asked:

“Food import from abroad has been increasing in the past. What is your opinion on this?”

40.4% of respondents consider it to be “bad” or “very bad”, while only 13.7% consider it to

be “good” or “very good.” 45.9% of respondents chose “can’t say one or the other.”

12

In 2003, average applied tariff rates for Japan, the U.S. and E.U. are 2.9%, 3.6%, 4.1%

respectively. 13 World Values Survey 1995.

8

Collective Action and Political Mobilization

The majority of literature on OECD countries attributes the high levels of

agricultural protectionism to the collective action capacity of producers over diffused and

unorganized consumers.14

Thus this literature has tested how the economic characteristics

of agricultural industries, such as the degree of geographic concentration, size of the

workforce, and import substitutability, affect their collective action capacity and thus the

level of protection they receive. The collective action argument, however, is unable to

speak to the fact that over half of the non-farmers, which constitutes 94% of the total labor

force in Japan and 96.1% of our nationally-representative survey respondents, support

protectionism.

Another strand of literature suggests that political mobilization by elites interact

with economic characteristics of industries to affect the levels of protection.15

Rogowski

and Kayser demonstrate that among OECD countries, countries with majoritarian systems

have lower prices for consumer goods than those with proportional representation systems.

This is due to low vote-seat elasticity which encourages organized interests like producers

to lobby harder under PR systems than under majoritarian systems. Park and Jensen have

also demonstrated that electoral systems that encourage legislators to target narrow

constituencies have higher levels of agricultural protectionism.

14

Gawande and Hoekman 2006; Mulgan 2000; Olson 1965, 1985; Park and Jensen 2007;

Rogowski and Kayser 2002. 15

Kabashima 1984; Magee, Block and Young 1989; Thies 1998; Grossman and Helpman 1994;

Sheingate 2001; McGillivary 2004; Park and Jensen 2007.

9

Similarly, the literature on Japan has blamed the multi-member district system

with single-non-transferrable-vote (SNTV) before 1996 as a source of high agricultural

protectionism.16

According to this view, the LDP politicians exchanged votes for

agricultural subsidies and protection especially in rural areas.17

Scholars have also argued

that the electoral reform of 1994 in Japan, which combines a single-member majoritarian

system with proportional representation might empower diffused interests like consumers

and decreases the power of organized interests.18

Using three waves of interest group

surveys since 1979, for instance, Kume demonstrates that special interest groups have lost

power over legislators after the reform.19

Indeed, if we look at legislators’ policy-positioning on agricultural policy, the

LDP politicians have become less responsive to farm interests over time. The all

Lower-House legislator survey we conducted during June-July of 2009, a month before the

general election, asked whether food import should be regulated (kisei suru). 55% of LDP

legislators oppose the additional limits on food import. The LDP decreased its support

for farmers around mid-1990s advocating for “agricultural policy in the offensive (seme no

nousei)” and identifying themselves as the “pro-globalization faction of agricultural tribe

politicians (kokusai-nousei-ha).”20

Indeed, the actual level of agricultural protectionism

16

Horiuchi and Saito 2009. 17

Scheiner 2006. 18

For an objection, see Bawn and Thies 2003. 19 Kume 2007.

20 Sasada 2009; Krauss and Naoi 2009. The LDP politicians who specialized their career

development in the agricultural sector became more responsive to food processing and fast food

10

measured by Consumer Support Estimates has gone down moderately from 51% in 1997

(post-electoral reform) to 41% in 2008.21

Despite this shift in elite attitudes as well as the level of protection, the public’s

support for agricultural protectionism has thus far survived the global recession. In sum, the

two arguments, collective action capacity and political mobilization, which focus on

producer interests and power, do not account for why consumers support agricultural

protectionism. To make sense of this puzzle, we need to question a widely-held

assumption that consumers are free traders.

Consumer Preferences in the Global Economy

The puzzle posed in the previous section calls us to move beyond the standard

theories of trade policy preferences, such as Stolper-Samuelson and Ricardo-Viner models.

These trade theories only focus on producer (or occupation-related) attributes of individuals.

Compared to the accumulation of occupational theories of trade policy preferences,

however, we still know little about consumer preferences on trade policy. This is because

the majority of literature has assumed consumer preference for free trade.22

The emerging literature, however, suggests various parameters beyond price

sensitivity that make some consumers more protectionist than others: the type of

chain industries. This might be due to their large financial contributions as food industries are one

of the fastest growing industries in terms of revenue and the number of employees. 21

OECD 2009. 22

Grossman and Helpman 1994.

11

consumption basket,23

safety and quality concerns,24

ethical concerns,25

the

love-of-variety,26

and community and family concerns.27

Ehrlich, for instance, finds that

the U.S. citizen’s support for fair trade products are not “protectionism in disguise” but

rather sincere concerns for labor and environmental standards. Hiscox and Smyth indeed

confirm in their field experiment at ABC Carpet that American consumers are willing to

pay extra dollars to purchase goods that are labeled to have met the labor standards.

In the Japanese context, Maclachlan has shown that although Japanese consumers

have demanded lower prices, as seen in the color-TV boycott incident in 1970, the major

policy goals of consumer organizations have been to secure consumer protection and the

right to choose (antimonopoly law). Vogel posits that Japanese consumer preferences might

not appear to be economically rational, as they did not demand neither trade liberalization

nor lower price. However, Japanese consumers demanded better quality and safety

regulations and were successful in obtaining them.28

Garon and Maclachlan demonstrate

that Japanese consumerism has had elements of “morality, thrift, community and national

identity” partly due to elites molding the public minds though government policies.29

23

Baker 2004; 2006. 24

Vogel 1999; Kono 2006. 25

Ehrlich 2009; Hiscox and Smyth 2009. 26

Krugman 1991; Broda and Weinstein 2007. 27

Goldstein, Margalit, and Rivers 2009. 28

Vogel argues: “to establish a causal link from group interests to policy outcomes we need to

determine that groups actually bore the assumed preferences, that they acted upon these preferences,

and that their actions influenced outcomes. Establishing a causal link requires empirical

verification.” 29 Vogel 1999; Garon and Maclachlan 2006; Maclachlan

12

Indeed, the most recent public opinion survey conducted by a Japanese

Think-tank suggests that Japanese citizens found the Chinese poisoned dumpling incident30

“more worrisome” than the world financial crisis and the political instability of North

Korea.31

If these food safety incidents shape Japanese citizens’ attitudes toward food

import, we expect that safety and quality concerns increase the public’s opposition to food

imports. In sum, the emerging literature on consumers has shown that their preference for

free trade is indeterminate at best: there are other parameters beyond price sensitivity that

might make consumers protectionist.

Why Experiment? Priming without Framing

The complexity of consumer preferences poses several major issues that have stalled

inquiry into how producer and consumer interests shape trade policy.32

The first is the

dual – and often conflicting – perspectives citizens have toward globalization as producers

versus consumers. Because the majority of citizens engage in both production (i.e.,

income-earning) and consumption activities, a key question is not whether consumers’

interests matter more than producers’; rather, we need to ask whether citizens’ support for

30

Japanese citizens have experienced a series of food-safety crises such as mad-cow disease in the

1990s and the Chinese poisoned dumpling incident in 2008. Japan Finance Corporation conducted a

series of surveys before and after the story about poisonous Chinese dumpling import dominated the

media. The surveys show a very sharp rise in safety consciousness and domestic food orientation

after the incident. Japan Finance Corporation, Shouhishadoukouchousa, 8/21/2009. Available at

http://www.afc.jfc.go.jp/topics/pdf/topics_090812a.pdf (last accessed, July 2009). 31

Mitsubishi Souken 2008. 32

Goldstein, Margalit and Rivers 2009; Author(s) 2009.

13

agricultural protectionism differs when they assess their positions in the global economy as

producers or consumers.33

[Chart 1 around here] This question calls for an experimental

research design that randomly primes citizens to think about globalization as a producer or

as a consumer.

Second, in light of studies that suggest consumers’ preferences are indeterminate and

complex, we should not frame respondents to think about the positive or negative

consequences of food import when we ask their attitudes toward it. For instance, Hiscox’s

framing experiments gave a different introductory statement to randomly assigned

treatment groups: “Many people believe that increasing trade with other nations creates

jobs and allows Americans to buy more types of goods at lower prices” or “Many people

believe that increasing trade with other nations leads to job losses and exposes American

producers to unfair competition.”34

After these introductory statements, the survey

proceeded to ask their support for free trade. In light of the complexity of consumer

preferences, however, this framing potentially poses a double-barreled problem as it

simultaneously primes respondents to think about trade from consumer vs. producer

perspective and frames them to think of trade’s positive or negative distributional

consequences. Instead, we need to design an experiment that primes respondents to think

33

There has been a long tradition of conceptualizing producer vs. consumer interests as a

dichotomy and trade-off in the political economy literature. See Smith 1776. 34

Hiscox 2006.

14

about food import from a consumer vs. producer perspective without framing its positive or

negative distributional consequences.

Research Design and Method

We conducted an on-line survey experiment in Japan with a sample of 1200

respondents between the ages of 20 and 65 during the first week of December, 2008 when

media coverage of the world financial crisis and the rise of unemployment among the

temporary workers was extensive.35

To understand the sources of support for agricultural

protectionism, we randomly assigned visual stimulus to three experimental groups. The

experiment consists of two groups that receive the treatment (“stimulus”) and another

control group without any stimulus (400 respondents each).

The producer-priming group is shown three photographs—a typical white-collar

office, a car factory, and rice field. The images were chosen so that they represent three

major sectors of the economy (service, manufacturing and agriculture) to activate

respondents’ consciousness as producers (or, their occupational interests) [Group 1 Photos

around here]. We chose images of typical workplaces with both male and female workers

present, except for the car factory image where all the workers appear to be male.

35

To ensure the representativeness of our sample, we also asked exact same questions on attitudes

toward food import in a nationally-representative public opinion survey in Waseda University’s

GLOPE with 1078 respondents (47.4% response rate) conducted during February of 2009. See the

discussion of results on pages 33-34 of this paper. The distribution of attitudes in our control group

are similar to the nationally-representative sample as shown in the Appendix. The designs and

results of two pilot studies we conducted before this survey experiment are available upon the

editor’s and reviewers’ requests.

15

The consumer-priming group is shown three photographs—a supermarket with

food, a consumer electronics retail store, and a large-scale casual clothing store. These

images encompass three areas of basic consumer goods that citizens purchase regularly

regardless of their income, gender, family status, and age.36

These visual stimuli are

intended to activate respondents’ consciousness as consumers [Group 2 Photos around

here]. The control group receives no stimulus. The treated and control groups are

balanced in their key demographic characteristics such as age, gender, income and skill

specificity as shown in the Appendix Table A-1.

In order to ensure that respondents indeed received the treatment,37

we asked a

follow-up question immediately after showing each image. For example, after showing

the picture of car factory, we asked: “What type of car do you think they are producing?”

Respondents choose from five options such as a racing car, a hybrid car, or a regular car.

After showing the photograph of supermarket, we asked, “What type of grocery shop do

you think this is?” in which respondents choose from five options ranging from a large

supermarket, to organic and natural food store, to small mom-and-pop shops. The idea is

not to test the facts about these images, but rather to ensure that respondents see these

photos and think of them either from job-related or consumption-related perspectives.

More than 70% (for the white-collar office photo) to 96% (for the rice field image) of

36

For instance, the image #2 is taken at the casual chain clothing shop called “Uniqlo” which is

culturally equivalent to the GAP in the U.S. At the end of the survey, we asked respondents whether

they own anything made by Uniqlo and more than 82.5% of respondents in both genders and

different age groups say Yes. 37

Horiuchi et al. 2006.

16

respondents chose correct answers to these questions confirming their reception of the

stimuli.

Using images to prime respondents has two advantages over framing experiments

which supply respondents with opinions about how trade affects consumers and

producers.38

First, priming differs from framing in that the former makes some issues

more salient than others and thus influences “the standards by which governments,

presidents, policies, and candidates for public office are judged,”39

while framing

characterizes issues negatively or positively.40

This characteristic of priming allows us to

manipulate respondents’ “standards” by which food import is judged (i.e., producer vs.

consumer) without imposing on them the judgment itself (i.e., food import is good or bad

for producers/consumers). This is critical for the purpose of our study as we do not yet

know whether activation of a consumer perspective uniformly leads to lower or higher

support for agricultural protectionism. Instead, the visual stimuli simply prime

respondents to think of themselves as consumers vs. producers.

Second, our visual stimuli do not explicitly convey information either about trade or

globalization. This is appropriate for the purpose of our study, as not all production and

consumption activities are linked, in reality and in citizens’ minds, to trade or globalization.

38

Hiscox 2006. 39

Iyengar and Kinder, 1987: 63. 40

Scheufele 2000.

17

After the treatment, we proceeded to ask attitudinal questions about food import and

general trade issues.41

The survey instruments are described in the results section below.

The Results: Aggregate Effects of Priming

General Trade Issues vs. Food Import

Table 1 summarizes the distribution of responses for a generic trade question,

“Import from foreign countries has been increasing in the past. What is your opinion on

this?” Respondents choose answers from a five-point scale (very good, good, neutral, bad,

and very bad). Consumer-priming substantially lowers the opposition to free trade: the

proportion of protectionist responses within the consumer-priming group is 13 percentage

points lower than that of the control group and 18 percentage points lower than the

producer-priming group. These differences are statistically significant at 5% level.

Producer-priming, on the other hand, turns out to have no systematic effects – while it

slightly increases the proportion of protectionist responses (by 4 percentage points over the

control group) the difference is statistically insignificant. Overall, the results suggest that

Japanese citizens are much more supportive of free trade when they assess their positions

on trade from a consumer perspective.

Table 2 summarizes the distribution of responses for the question, “Food import from

foreign countries has been increasing in the past. What is your opinion on this?” in which

41

The original Japanese language version is available at the editor’s and reviewers’ request.

18

respondents choose answers from the same five-point scale.42

Table 2 first suggests that

respondents’ opposition to food import is much stronger than that of general import of

goods. Among all three experimental groups, the proportion of protectionist responses is

at least 20 percentage points higher for food import compared to general imports (Graph 1).

Contrary to our expectation, however, consumer-priming does not provoke higher

opposition to food import. Rather, producer-priming increases the proportion of

protectionist responses by ten percentage points compared to the control group. The

difference is statistically significant at the 5% level. [Figure 3 around here] This is

puzzling as an increase in food import should not directly affect the job security and wages

of the respondents except for a small fraction of farmers (0.7% in our sample; 3.9% in

GLOPE survey; 6% of total workforce in Japan including part-time and weekend

farmers).43

This finding poses two puzzles. First, why does the activation of a producer

perspective lead to higher support for agricultural protectionism? Second, this finding

forces us to reconsider the “protectionist consumers” argument made by Vogel and others.

If Japanese citizens support agricultural protectionism due to concerns over the safety and

42

Note that this question asks respondents’ opinion about “increasing food import” not about

protecting agriculture or farmers. We chose this form of question as it does not directly remind

respondents about “jobs” (theirs or farmers’) or “consumption”. Given this neutrality, it is even

more puzzling that the producer-priming increases the opposition to “food import”. 43

On the other hand, those who are employed in food-processing and restaurant/fast-food chain

industries will benefit from increasing food import. Indeed, the Japanese fast-food chain

association has extensively lobbied the LDP legislators for opening of the agricultural market and

accepting more foreign workers.

19

quality of imported goods (or, food nationalism), consumer-priming would have made them

more protectionist. But for the general trade issue, consumer-priming is associated with

much higher support for free trade and it had no systematic effects on citizens’ attitudes

toward food import. If quality and safety concerns do not dictate the public’s support for

agricultural protectionism, what does?

Disentangling the Puzzling “Consumers think like Producers”

We disentangle this puzzle along two possibilities suggested by experimental

studies on mass support for income redistribution: sympathy of the public for “poor” and

“hardworking” people (in our case, farmers), and the public’s projection of their own job

insecurity onto a symbolic declining industry (i.e., agriculture).44

Both mechanisms force

us to move beyond occupational theories of trade policy preferences, based on

Stolper-Samuelson and Ricardo-Viner models, which derive individual policy preferences

from their relative positions in the international economy. Instead, we should consider

how individuals perceive other occupations or sectors (i.e., agriculture) when forming their

44

Another, often-cited reason for the public support for agricultural protectionism is a form of

nationalism (food nationalism). We tested the effect of nationalism in two ways and found no

effects. First, we asked respondents what their main identity is and included in the models a

dummy variable indicating one for those who chose “nationality” or “ethnicity” as one of the top

two identities they have, and zero otherwise. Not only was the proportion of “nationalist”

respondents quite small at 3.9%, the nationalism dummy turns out to have no effect. Second,

when we asked respondents’ levels of exposure to protectionist elite discourse (discussed in details

on page 26), we included four terms that are directly related to food nationalism to the protectionist

information index: food self-sufficiency ratio, food education, locally-grown locally-consume, and

Chinese poisoned dumpling incident. This variable, Protectionist_Info, turns out to have no

systematic effects.

20

own attitudes toward trade.45

To do so, we analyze which subgroups of respondents are

sensitive to producer and consumer-priming and identify the direction of their attitudinal

differences among the three experimental groups.46

Sympathy

One possible explanation for the puzzle is that producer-priming provokes

agricultural protectionism due to the sympathy that consumers have about farmers: their

occupational images are hardworking, but declining with low pay. Indeed, despite the fact

that the household income of farmers exceeded that of the average employee since 1975,47

sociologists have found that farmers’ “occupational prestige scores” in which citizens rank

the socio-economic prestige of more than 80 occupations since 1955 has been extremely

stable and low throughout 1990s.48

Citizens might perceive agricultural protectionism as

a redistributive policy in the context of post-War Japanese political development.49

45

Mansfield and Mutz 2009 demonstrate that the U.S. citizens form their attitudes toward trade

socio-tropically rather than individually due to the exposure to media and elite discourse on trade’s

effect on the national economy, as opposed to on their individual occupation. Our approach

echoes their view in that it deals with social determinants of individual attitudes, yet differs from

them by theorizing how citizens’ perceived similarities and differences with other sectors can

determine their attitudes toward protecting a given sector. 46

One cannot infer the effect of stimuli on respondents by simply looking at aggregate distribution

of responses (“average treatment effects”) when the heterogeneity of treatment effects exists across

sub-groups (“local treatment effects” e.g., consumer-priming increases support for food import for

men and reduces it for women). Angrist 2004; Imai and Strauss 2009; Bitler, Gelbach and Hoynes

2007. 47

Moriguchi 2008. 48

Naoi 1979; Hara 1999. The occupational prestige score for farmers (jisakunou) is 51, 43, 45, 46

for years 1955, 1965, 1975, and 1995 respectively. These scores are comparable to taxi drivers,

workers at train stations (ekiin), and hair stylists (riyoushi) but much lower than white-collar

employees. 49

Kabashima 1984; Calder 1988. Since the mid-1950s when the Japan’s high economic growth

21

Experimental studies on people’s attitudes toward redistribution also lend support

to this intuition. While the rationalist model expects that support for income redistribution

decreases as personal income increases, the emerging survey evidence finds that a large

number of low-income respondents oppose redistribution and a large number of

high-income respondents support it. The level of respondents’ income, thus, is only a

partial predictor of attitudes toward redistribution. Survey research conducted by Fong

shows that people express stronger support for redistribution to the poor when they think

that poverty is caused by bad luck rather than laziness.50

Survey experiments on public

support for government assistance to natural disaster victims confirm this finding that

people support redistribution when they sympathize with the recipients and feel that they

“deserve” it.51

Fong elaborates this finding with an experiment. She designed a charity game,

in which a player is randomly matched with a real-life welfare recipient who exhibits strong

or weak work ethics. The player is then asked how much s/he is willing to contribute to the

recipient from the fixed amount of money the player is given. The study found that the

player gives more money to the welfare recipient with a strong work ethic. These studies

find that strong reciprocity better explains the motivations behind the welfare state:

started, both the Japanese and the U.S. governments’ concern was how to reduce the urban-rural

income gap as this gap was considered to be a root cause of Japan’s militarism during the WWII. 50

Fong 2001:285. 51

Harris-Lacewell et al. 2009.

22

generosity toward the poor is conditional on the belief that the poor work hard.52

Lu,

Scheve and Slaughter also formally demonstrate that altruism can be the reason why

low-skill and labor-intensive industries, such as agriculture, receive high levels of

protection across countries with different factor endowments such as the United States and

China.53

In order to identify whether sympathy is a source of support for protectionism,

we asked the respondents to choose three words each that characterize their images and

feelings toward producers and consumers.54

This was done before respondents received

the visual stimuli. Among twenty word choices,55

the top-four producer images chosen

were “responsibility” (“sekinin”, 48.3%), “sweat” (“ase”, 48.1%), “country/rural”

(“inaka”, 43%), and “factories” (“koujou”, 36.7%). On the other hand, the top-four

consumer images are “money” (“okane”) (57.9%), “citizens” (“shimin”) (46.0%),

“information” (“jouhou”) (28.6%) and “urban/metropolitan” (“tokai”) (27.7%).

Two points are worth highlighting. First, a higher proportion of respondents

chose “country/rural” (i.e., agriculture) over “factories” (i.e., manufacturing industries) as

52

Fong 2004:278-282; Williamson 1974; Heclo 1986. 53

Lu, Scheve and Slaughter 2009. 54

The question wording is: “We call those who produce manufactured and agricultural products as

well as those who provide service to customers ‘producers’ (seisansha), and call those who

purchase these goods and consume ‘consumers’ (shohisha)”. Among the 20 words below, please

each choose three images or feelings you have about producers and consumers.” 55

These are trust (sekinin), suspicion (utagai), urban (tokai), rural (inaka), money (okane), leisure

(goraku), responsibility (sekinin), information (jouhou), weekdays (heijitsu), off days (kyujitsu),

sweat (ase), factories (koujou), government (seifu), citizens (shimin), progressive (kakushin),

conservative (hoshu), men (dansei), women (josei).

23

their image of producers. Second, “money” is much more closely associated with

consumers (57.9%) than producers (6.8%). This means that the majority of respondents

have images of producers being hardworking (“sweat”) in rural areas or hardworking

blue-collar workers in factories without much monetary reward. Thus, we include a

variable Sweat which takes a value of one if a respondent chooses “sweat” as one of the

three producer images and zero otherwise.56

The second variable is the respondents’ attitudes toward redistribution

(Redistribution). We recode respondents’ five-scale responses to a question, “What is

your opinion about a policy to redistribute wealth from the rich to the poor using taxation

and the social insurance system?” so that the higher the number, the higher the support for

redistribution (e.g., 4: very high support, 0: very high opposition).

Farm-to-Table is equal to one when respondents have used a “farm-to-table”

service (“sanchoku”) to buy food directly from farmers in the past year, and zero otherwise.

Our expectation is that those who have used “farm-to-table” services might sympathize

with farmers more as they communicate with farmers regarding the products via mailing

lists or direct conversation. Finally, Social Network takes a value of one if a respondent

has a family member or relatives who engage in farming, including part-time farming.

Indeed, while only 0.7% of our respondents are farmers, 37.7% of respondents have a

family member or relatives engaging in farming. This variable is included to test a

56

Alternatively, we also include an interaction term between “sweat” and “rural”. The results do

not change.

24

popularly believed social network argument that consumers support agricultural

protectionism because their family members are engaged in farming.

Projection

The second hypothesis we test is that citizens might support agricultural

protectionism because they project their own job insecurities onto a symbolic declining

industry, agriculture. Projection is a concept developed in social psychology to

understand how people make inferences about ‘others’ using their own mental states as a

benchmark.57

Ames develops “projection” and “stereotypes” as two inferential strategies

people use to understand what others want. With lab experiments, he demonstrates that

when the perceived similarity between self (i.e, a perceiver) and others (i.e., a target) is

high, respondents are more likely to use projection as a tool to infer others’ preferences.

On the other hand, when the perceived similarity between self and others is low,

respondents are more likely to use stereotypes as a mechanism of inference.

In the context of our research, this means that when the level of respondents’

perceived similarity with farmers is high, respondents are more likely to project their own

mental states onto what farmers want regarding food import (i.e., protectionism). We derive

three potential similarities that respondents might perceive with agricultural workers:

declining industry, high job insecurity (i.e., difficulty finding a comparable job), and older

57 Freud 1936; Ames 2004 a, b.

25

age. In projection research, these perceived similarities are usually recorded before asking

respondents to infer the targets’ positions. We did not want to ask about respondents’

perceived similarities with farmers before our question about food import, however, due to

our concern that the similarity questions could risk priming respondents to think about

farmers. Thus, these potential similarities are derived from conventional occupational

images of farmers found in existing public opinion and social surveys.58

We aim to test

whether the subgroup of respondents who are likely to perceive potential similarity with

agricultural workers shows higher opposition to food import interacting with the producer

and consumer stimuli.

Thus, if the projection hypothesis holds, we expect to see respondents with high

job insecurity or perceived risk of income loss in the future show higher support for

agricultural protectionism. To test this, we include Skill Specificity which is a five-point

scale response to the following question: “If you were to quit your current job, do you

believe it would be difficult to find a similar job that pays a comparable salary?” We

recode this variable so that the higher the number, the harder it is for a respondent to find

their next job.

Likewise, older respondents face a higher risk of income loss due to

forthcoming retirement or increasing difficulty in finding their next job. Also, the average

age of agricultural workers in Japan is 57.6 for all the farmers and 64.6 for farmers whose

58

Hara 1999.

26

main source of income is farming.59

Age is a respondent’s biological age coded into one

of five categories (20-29, 30-39, 40-49, 50-59, 60-65). We trichotomize job status as

three dummy variables to FullyEmployed for full-time employees, Temp for part-time and

temporary employees, and No Jobs for the unemployed and those not in the labor force.

Fully Employed is excluded from the model as a base category.

Alternative Hypotheses and Controls

Building on the existing work on individual attitudes toward globalization in

general and food import in particular, we include a battery of covariates which capture

respondents’ job and consumer-related attributes.

For respondents’ job-related attributes, the following covariates are included.

Low Income indicates an individual in the a bottom 30% of the income distribution, High

Income indicates top 30% and Mid Income is the base category. College is a dummy

variable; it is equal to one for respondents who graduated from college and zero otherwise.

Female is a dummy variable; one for women and zero otherwise.

We also control for social and political attributes that might influence

respondents’ attitudes. Married is a dummy variable equal to one for married respondents

and zero otherwise. Have Kids is a dummy variable “1” for respondents with children and

zero otherwise. Housewives is a dummy variable “1” for housewives and zero otherwise.

59

Data for 2008. Japanese Statistics Office,

available at http://www.e-stat.go.jp/SG1/estat/List.do?bid=000001014189&cycode=0

27

LDP and Non-PartyID indicate respondents’ party identification with then-incumbent

Liberal Democratic Party and non-partisan, respectively.60

We also include an index

measure of political participation, Politically Active, which is the sum of the answers (0: not

at all, 1: once or twice, 2: several times) to three questions about the frequency of political

participation in the past: whether a respondent has been a member of a local candidate

support group (koenkai), whether a respondent has helped a candidate run for an election,

and whether a respondent has listened to politicians’ speeches on the street. Politically

Active ranges from zero to six.

We also include three novel covariates to remedy some issues found in the

existing literature. One variable is Protectionist_info, which measures respondents’ levels

of exposure to elites’ (politicians, bureaucrats, and the media) protectionist discourse on

trade policy. Generally, the level of respondents’ exposure to elite discourse is proxied

either by education or respondent’s answer to a quiz on technical aspects of trade policy.61

We attempt to directly measure the level of exposure by asking a series of questions about

whether they have heard and/or spoken with their colleagues or family about the following

terms: “Free Trade Agreement (FTA or EPA)”, “World Trade Organization (WTO)”,

60

We also tested alternative models of party IDs where LDP, DPJ, and non-partisan support are

included in the model and the remaining parties (mostly leftists, such as Social Democratic Party

and Japan Communist Party and center-left Clean Government Party) are the base category. The

results do not change. 61

Baker 2009. For instance, Baker quizzes respondents to list signatory countries for CAFTA for

his survey in Latin America. While Baker’s quiz gets at respondents’ sophistication regarding

trade issues and their levels of technical knowledge, our index attempts to measure the respondents’

levels of exposure to elite discourse.

28

“Safeguard provision”, “poisoned Chinese dumpling incident (chugoku doku gyoza jiken),”

“food education (shokuiku)”, “food sufficiency ratio (syokuryo jikyuritsu),” and

“locally-grown, locally consume (chisan chisho)”.62

The former three terms concern issues

related to high trade politics (e.g., bilateral or multilateral diplomacy) and the latter four

terms are extensively used by politicians and the media to either raise the barrier to trade or

promote domestic production and consumption of food and agricultural products.

The data suggests that one quarter of the respondents is highly exposed to

protectionist elite discourse but not to the high trade politics discourse: 24.75% of the

respondents have heard and spoken about all of the four protectionist terms. The variable

Protectionist_info is thus constructed as a dummy variable “1” if a respondent heard and

spoke about all of the four protectionist terms and zero otherwise in order to test how the

level of exposure to protectionist elite discourse mediates or exacerbates the effect of

priming.63

The second novel variable is No Foreign Transaction which takes a value of one

when the company with which the respondent is employed or the business run by the

62

We chose these terms by (i) examining the websites of members of parliament mentioning

“agriculture (nougyo)” or “farmers (nouka)” and identifying common reasons they mention to make

the case for protection; (ii) using questionnaires in the Ministry of Foreign Affairs’ public opinion

surveys on Economic Diplomacy (Keizai Gaikou no Ishiki Chosa) conducted in 2004; and (iii)

reflecting key legislation or policy (FTAs, Food Education Law in 2005) and events (poisoned

dumpling in 2008) during the past five years. We have requested the survey data from the

Ministry on numerous occasions since summer of 2007, yet have had no luck obtaining it thus far. 63

Alternatively, we included a variable High Trade Info which is an index of the number of terms

in high trade politics (“FTA”, “WTO”, and “safeguard provision”) a respondent has heard or spoken

with family or colleagues about. This variable, however, turns out to have no systematic effects.

29

respondent does not import, export or outsource its production activities abroad and zero

otherwise. This variable is included to remedy the criticism that the conventionally-used

ILO occupational classification is not an accurate approximation of skill-level or

export-orientation vs. import-competing sectors. The self-reported position of a

respondent’s company or sector might provide more accurate data on how respondents

perceive their job’s relative position in the global economy.64

Finally, we include several questions to capture consumer attributes of the

respondents which are generally missing in standard public opinion surveys. First, we list

the names of widely recognized luxury shops and discount stores in Japan and ask

respondents whether they own any product from each store. These stores are: Louis

Vuitton, Gucci, Armani, Prada, Channel, Brooks Brothers, Ralph Lauren, Rolex, Tumi,

Uniqlo, Muji, Gap, Ikea, Body Shop, and Aoki. A variable named Cheap_shopper is an

index of the number of items a respondent owns from the following discount stores (Uniqlo,

Muji, Gap, Ikea, and Aoki) and ranges from zero to five.65

64

Interestingly, 16.25% of the respondents (195 out of 1200) report they do not know whether their

company or sector imports, exports, or outsources production activities abroad. This finding

cautions that the link between respondents’ occupations and their attitudes toward globalization is

not as obvious as conventionally thought. 65

Alternatively, we ran the same model with a variable named “Brand Shopper” which is an index

of the number of stores a respondent owns a product from the following stores: Gucci, Armani,

Prada, Channel, Brooks Brothers, Ralph Lauren, Rolex, and Tumi. This variable turns out to have

no significant effects on the effect of priming. It is also worth reporting that these shopping habits

cannot be proxied by income. We ran a linear model with this Brand Shopper as the dependent

variable, and income, education, gender, age, and whether respondents’ jobs do not import, export,

or outsource (No Foreign Transaction) as covariates. High income and female respondents own the

highest numbers of items from these shops, yet the substantive impact of income is very small: a

shift from the mid income category (33%<n<67%) to the high income category (n>67%) only

30

The Models and Estimations

We pool all the data across the treatment and control groups and estimate the

treatment effects by interacting the treatment group dummy for each experiment group

(0-1) with each of the respondents’ attributes.66

The control group is the base category

which is compared to the producer- and consumer-priming groups. The model, estimated

by ordered logit, has the following structure where T1 is a dummy variable (“1”) for

producer-priming and zero otherwise and T2 is a dummy variable (“1”) for

consumer-priming:

Support for Food Import i= !0i + T1*!1 Sympathy/Projection i + T2*!1 Sympathy/Projectioni + T1*!2 Job-related attributes i + T2*!2 Consumer attributes i + T1*!3 Protectionist Info i + T2*!3 Protectionist Info i + T1*!4 Political Mobilization i + T2*!4 Political Mobilization i

Results (2): Sympathy and Projection

Table 3 summarizes the results of an ordered logit analysis of which respondents

are sensitive to priming. A positive coefficient on a variable indicates that with producer

or consumer priming, a given subgroup’s support for food import increases. Negative

coefficients indicate that, again, interacting with the producer or consumer priming, a given

increases the number of items owned among these shops at 0.6 within a range of zero to eight.

66

All the covariates, except for “Protectionist_info” are individual attributes that cannot be

changed due to the priming. One could argue that the priming might affect responses through

changing the intermediate variable (“mediation effects”, see Imai, Keele and Yamamoto 2009) such

as “Protectionist_info.” The distribution of responses to “Protectionist_info” does not differ,

however, across the treatment and control groups. The presence of mediation effects are therefore

not warranted here.

31

subgroup’s support for food import decreases. Models 1 and 2 use a five-point scale (very

good, good, neutral, etc) as the dependent variable and estimates it with an ordered logit,

while Model 3 dichotomizes the five-point scale to zero (opposition to food import; “bad”

and “very bad”) and one (support for food import including the neutral responses) and

estimates it with a binomial logit.67

Overall, the results of our experiment lend strong support to the projection

hypothesis and very weak support for the sympathy hypothesis. With producer-priming,

respondents with high job insecurity (Skill Specificity) and higher biological ages (Age) are

the ones who become more protectionist (i.e., higher opposition to food import). With

consumer-priming, those who are not employed (No Jobs) increase their support for food

import which is consistent with our expectation that unemployed respondents prefer lower

prices. This result is robust across the three models.

Figure 4 shows the substantive impact of respondents’ self-assessed job

insecurity on their decision to support agricultural protectionism. As skill specificity

increases from “easy” to “very difficult,” the proportion of protectionist responses increases

from 37.0% to 68.8% in the producer-priming group. Moreover, Skill Specificity has no

systematic effects on the proportion of protectionist responses in the consumer-priming and

67

For the sake of comparison, we also present Table A-2 and A-3 in the appendix which shows the

results of the same ordered logit and binomial logit models only with a control group in our

experiment and with a nationally-representative public opinion survey conducted by GLOPE in

February 2009. This gives us benchmark variations of individual support for food import among the

subgroups without treatment (“priming”) which will be discussed in detail later.

32

control groups which provides more evidence for the projection hypothesis.

On the other hand, the sympathy hypothesis finds only very weak support. Only

in Model 3 with dichotomized support for food import (0-1), with producer-priming, do

respondents with a family member or relatives engaging in farming (Social Network) show

higher opposition to more food import than a control group. Other variables for the

sympathy hypothesis, such as Sweat and Farm-to-Table have no systematic effects.

Respondents’ attitudes toward redistribution (Redistribution) have the opposite of the

expected effect: higher support for government redistribution is associated with higher

support for food import, especially with producer-priming.

A possible interpretation of this puzzle is that the Japanese public may be aware

that two forms of income inequality exist, class and urban-rural. Agricultural protectionism

is all about the redistribution of income from urban to rural at the expense of redistribution

from high to low income. Kume’s survey on interest group leaders lends additional support

to this perceived trade-off: labor union leaders are much less concerned about regional

inequality than other interest groups, despite the fact that leaders consider income

inequality to be a serious concern.68

Alternative hypotheses and control variables find partial support. With

producer-priming, respondents with the top 30% of income (High Income) are more

supportive of food import than a control group. Other job-related covariates, such as

68

Kume 2006.

33

college degrees and whether a company or sector of employment exports, imports, or

outsources production abroad, turn out to have no systematic effects. On the other hand,

consumer-priming makes respondents with children more supportive of food import. This

is probably driven by the price sensitivity. The consumer-priming, however, makes

married respondents more protectionist, controlling for income and whether they have

children or not. The finding is consistent with Goldstein, Margalit and Rivers who argue

that married respondents are more protectionist, yet provides an alternative explanation that

this might have to do with their attitudes toward consumption, rather than occupational

interests.69

Co-op members become more protectionist with consumer-priming which is

expected.

LDP supporters become more supportive of food import with producer-priming.

This is contrary to the conventional wisdom that the LDP is “pro-farmer”, but the finding is

in fact consistent with the preliminary finding of our Lower-House legislator survey that

LDP politicians are more pro-globalization and pro-food import than the then-opposition

Democratic Party of Japan’s politicians. Contrary to the expectation of the political

mobilization hypothesis, respondents’ level of exposure to elites’ protectionist discourse

(Protectionst_Info) and their political activeness (Politically Active) turn out to have no

systematic effects.

Table A2 and A3 in the appendix show the results of the same analysis only with

69

Goldstein, Margalit, and Rivers 2009. They show that married respondents are less risk averse

and that their attitudes toward risk accounts for the higher level of support for protectionism.

34

a control group in our experiment and with a nationally-representative GLOPE survey.

The benchmark results suggest that without producer-priming, neither a sympathy or

projection mechanism holds, except for a weak support for the projection hypothesis found

in Model 4 of Table A3.70

This lends additional support to our experimental finding that the

producer-priming is what encourages respondents to link their own job insecurity with that

of farmers. Without such stimuli, respondents generally do not project their job insecurity

onto a symbolic declining sector, agriculture.

Indeed, within a nationally-representative survey sample, temporary workers

(Temp) are more likely to support food import than fully-employed respondents which is

consistent with their expected economic needs for cheaper food. Given the higher baseline

support for food import among temp in the nationally-representative survey, our

experimental finding that temp became more opposed to food import than fully-employed

respondents with producer-priming is paradoxical. Although our findings are confined to

unraveling the mechanism for mass support for agricultural protectionism (i.e., internal

validity), this finding helps us solve the puzzle of why mass support for agricultural

protectionism has increased over time during the Japan’s decade long recession followed by

the world-wide financial crisis.

Conclusion

This paper has aimed to make three contributions. First, our study brought

consumers and the mass public to the center of the analysis of agricultural protectionism.

70

Skill specificity is associated with lower support for food import although at 90% level

significance.

35

The collective action literature argues that organized powerful producers (farmers)

successfully pursue protection at the expense of diffused consumers. This assumption has

become a pillar of the majority of studies on agricultural protectionism. We challenge this

very assumption by first systematically presenting the puzzle that over half of the public in

Japan and elsewhere in fact support agricultural protectionism even during the global

recession. We then disentangled various sources of this high support with a randomized

survey experiment and demonstrated that the high level of mass support is partly due to

respondents’ projection of their own job insecurity onto a symbolic declining industry,

namely agriculture.

The finding is paradoxical as people with high job insecurity should be the prime

beneficiaries of agricultural liberalization as consumers. This paradoxical finding forces

us to reconsider standard theories of trade policy preferences which are based on the

premise that individuals choose the policy positions that best fit their occupational interests

in the international economy. Beyond the issues of trade and globalization, this finding

suggests that we should reconsider how individuals perceive “self interest” and infer

“interests of others” to better understand mass support for protectionism and redistribution.

Second, the projection finding also may serve as an important step toward

understanding how a coalition of diverse interests emerges in the politics of globalization

and redistribution - a missing link between individual preferences and coalitional politics in

the current research on public opinion in IPE. Because employees in declining sectors

36

perceive similarities with agricultural workers, non-farmers support agricultural

protectionism which in turn harms their interests as consumers.71

Rather than attributing

this paradox with ideological or cultural pre-disposition of consumers, we have

demonstrated the mechanism of projection which gives rise to new coalition between

producers and consumers. Our finding that there is an emerging alliance of producers and

protectionist consumers will push forward the long-standing research on coalition politics

in the global economy which mostly predict class or sectoral alliances. For instance, this

coalition of economic losers (i.e., workers in declining sectors) and farmers differs from

“the marriage of iron and rye” during Bismarck Germany (1871-1914) where the

government protected both landowners and capitalists at the expense of workers

(consumers) by politically sustaining the high price of grains.72

Third, using the “priming without framing” survey experiment, we have

demonstrated that citizens’ attitudes toward globalization and trade can differ dramatically

depending on which aspect of their lives (work vs. home and social lives) is activated from

their multifaceted and often conflicting attributes. A promising line of future research,

thus, is to investigate how elites – legislators, bureaucrats, and media –seek to activate the

job vs. consumption-related interests of voters and which groups of voters are susceptible

to such elite priming and framing. The general election of 2009 in Japan might provide an

71

Peter Linderdt 2004 has coined the term “social affinity” to explain the mass support for

redistribution in OECD countries. Our projection hypothesis is similar to his but goes beyond it

by deriving perceived similarities between the subject (respondents) and the object (farmers). 72

Rogowski 1989.

37

opportunity to explore this question as the major opposition party, the Democratic Party of

Japan, has extensively campaigned to appeal to consumers and improving their quality of

lives (“seikatsu” and “kurashi”), while the LDP has focused its campaigns on “creating

jobs.”

Another promising line of research is to explore the external validity of our

projection finding by bringing this experiment to racially and ethnically diverse societies

such as the U.S. or India.73

Along the line of observational studies on racial diversity and

income redistribution in U.S. cities by Alesina and Baqir, we expect that the projection

mechanism is prevalent in more homogenous societies (e.g. Japan) than heterogeneous

societies (the U.S. and India).74

74 Alesina and Baqir 1999.

38

Bibliography

Ames, Daniel R. 2004a. Strategies for social inference: A similarity contingency model of

projection and stereotyping in attribute prevalence estimates. Journal of Personality and

Social Psychology, 87, 573-585.

Ames, Daniel R. 2004b. Inside the mind-reader’s toolkit: Projection and stereotyping in

mental state inference. Journal of Personality and Social Psychology, 87, 340-353.

Baker, Andy. 2005. Who Wants to Globalize? Consumer Tastes and Labor Markets in a

Theory of Trade Policy Beliefs. American Journal of Political Science, Vol. 49, No. 4

Baker, Andy. 2009. The Market and the Masses in Latin America: Policy Reform and

Consumption in Liberalizing Economies. New York: Cambridge University Press.

Bauer, Raymond A, Ithiel de Sola Pool, and Lewis Anthony Dexter. 1972. American

Business and Public Policy: The Politics of Foreign Trade, 2nd ed, Chicago:

Aldine-Atherton.

Bawn, Kathleen and Mark Thies. 2003. A Comparative Theory of Electoral Incentives

Representing the Unorganized Under PR, Plurality and Mixed-Member Electoral Systems.

Journal of Theoretical Politics, Vol. 15, No. 1

Beaulieu, Eugene. 2002. Factor or Industry Cleavages in Trade Policy? An Empirical

Analysis of the Stolper-Samuelson Theorem. Economics and Politics 14(2):99–131.

Burgoon, Brian, and Michael J Hiscox. 2003. The Mysterious Case of Female

Protectionism: Gender Bias in Attitudes Toward International Trade, Paper presented at the

Annual Meeting of the American Political Science Association, August, Philadelphia

Broda, Christian and David E. Weinstein. 2004. Variety Growth and World Welfare. The

American Economic Review, Vol. 94, No. 2

Davis, Christina. 2003. Food Fights over Free Trade: How International Institutions

Promote Agricultural Trade Liberalization, Princeton: Princeton University Press.

Davis, Christina. 2004. International Institutions and Issue Linkage: Building Support for

Agricultural Trade Liberalization. American Political Science Review, 98-1.

Ehlrlich, Sean. 2009. The Fair Trade Challenge to Embedded Liberalism, Manuscript,

Florida State University.

39

Fong, Christina M. “Social Preferences, Self-Interest, and the Demand for Redistribution,”

Journal of Public Economics 82, 2. 2001: 225-246.

Fong, Christina M. 2007. “Evidence from an Experiment on Charity to Welfare Recipients:

Reciprocity, Altruism and the Empathic Responsiveness Hypothesis.” Economic Journal,

117.522, 2007 July.

Garon, Sheldon and Patricia L. Maclachlan (Ed). 2006. The Ambivalent Consumer:

Questioning Consumption in East Asia and the West. Cornell University Press.

Gawande, Kishore and Bernard Hoekman. 2006. “Lobbying and Agricultural Trade Policy

in the United States”, International Organization, 60:3:527-561.

Gintis, Herbert, Ammuel Bowels, Robert Boyd and Ernst Fehr, eds. 2005. Moral

Sentiments and Material Interests: The Foundations of Cooperation in Economic Life, The

MIT Press, 2005.

Goldstein, Judith, Yotam Margalit and Douglas Rivers. 2008. “Producer, Consumer,

Family Member: The Relationship between Trade Attitudes and Family Status” Paper

Prepared for the Princeton Conference on Domestic Preferences and Foreign Economic

Policy, Princeton University, April 19-20, 2008.

Grossman Gene M. and Elhanan Helpman. 1994. “Protection for Sale” American Economic

Review Vol. 84, No. 4, pp. 833-850

Hainmueller, Jens and Michael J. Hiscox. 2006. Learning to Love Globalization: The

Effects of Education on Individual Attitudes towards International Trade, International

Organization, Vol. 60, No. 2 (Spring, 2006): 469-498.

Hara, Junnosuke. Ed. 1999. Shakai Kaiso: Yutakasa no naka no fubyodo [Social

Stratification: Inequality in Wealthy Society]. Tokyo: University of Tokyo Press.

Harris-Lacewell, Melissa, Kosuke Imai, and Teppei Yamamoto. 2009. “Racial Gaps in the

Responses to Hurricane Katrina: An Experimental Study,” Working Paper.

Heclo, Hugh. 1986. “The Political Foundations of Antipoverty Policy.” In Sheldon H.

Danziger and Daniel H. Weinberg (eds.), Fighting Poverty: What Works and What Doesn’t,

Harvard University Press, 1986, 312-341.

Hiscox, Michael J. 2002. International Trade and Political Conflict. Princeton: New

Jersey: Princeton University Press.

40

Hiscox, Michael J. 2006. Through a Glass and Darkly: Framing Effects and Individuals’

Attitudes towards International Trade, International Organization, Vol. 60, No. 3 (Summer,

2006): 755-780.

Hiscox, Michael J. and Nick Smyth. 2009. Is There Consumer Demand for Improved Labor

Standards? Evidence from Field Experiments in Social Labeling. Manuscript, Harvard

University.

Horiuchi, Yusaku and Jun Saito. 2009. Cultivating Rice and Votes: Demand-Side Origins

of Agricultural Protectionism in Japan. Working Paper.

Imai, Kosuke, Luke Keele, and Teppei Yamamoto. 2009. “Identification, Inference, and

Sensitivity Analysis for Causal Mediation Effects”. Working Paper.

Iyengar, Shanto. 1991. Is Anyone Responsible? How Television Frames Political Issues.

Chicago: University of Chicago Press.

Jacoby, William G. 2000. Issue Framing and Public Opinion on Government Spending+

American Journal of Political Science 44 (4):750 – 67.

Kabashima, Ikuo. 1984 “Supportive Participation with Economic Growth: The Case of

Japan” World Politics, Vol. 36, No. 3 (Apr., 1984), pp. 309-338.

Kahneman, Daniel, and Amos Tversky. 1979. Prospect Theory: An Analysis of Decision

Under Risk. Econometrica 47(2):263–91.

Kono Daniel Y. 2006. Optimal Obfuscation: Democracy and Trade Policy Transparency.

American Political Science Review, 100-3.

Kume, Ikuo. 2006. Roudou Seiji [Labor Politics]. Tokyo: Chuou Kouron Shuppansha.

Kume, Ikuo. 2007. "How Special Interest Group Politics Falls: The Case of Japan" Paper

presented at the Conference, "Modeling Power Relationship in Japanese Democracy"

University of British Columbia August 28-29, 2007

Krugman, Paul R.(1979), Increasing returns, monopolistic competition and international

trade. Journal of International Economics vol. 9.

Linderdt, Peter. 2004. Growing Public: social spending and economic growth since the

eighteenth century. Cambridge University Press.

Lu, Xiaobo, Kenneth F. Scheve, and Mathew Slaughter. 2009. Protecting the Poor: Skill

Bias in the International Distribution of Trade Protection. Working Paper.

41

Lupia, Skip and Mathew D McCubbins. 1998. The Democratic Dilemma: Can Citizens

Learn What They Need to Know? New York: Cambridge University Press.

Maclachlan, Patricia L. 2001. Consumer Politics in Postwar Japan: The Institutional

Boundaries of Citizen Activism. Columbia University Press.

Magee, Stephen P. and William M. Brock and Leslie Young. 1989. Black Hole Tariffs and

Endogenous Policy Theory. MA. Cambridge University Press.

Mansfield, Edward and Diana Mutz. 2009. Support for Free Trade: Self-Interest,

Sociotropic Politics, and Out-group Anxiety. International Organization. Vol.63, Issue 3.

McGillivray, F. 2004. Privileging Industry: The Comparative Politics of Trade and

Industrial Policy. Princeton University Press.

Mulgan, Aurelia George. 2000. The Politics of Agriculture in Japan. London: Routledge.

Naoi, Atsushi. 1979. “Shokugyoteki Chiishakudo no Kosei (The Construction of the

Occupational Status Scale)” in Ken'ichi Tominaga (ed.), Nihon no Kaiso Kozo, Tokyo:

Tokyodaigakushuppankai.

Naoi, Megumi and Ellis Krauss. 2009. Who Lobbies Whom? Special Interest Politics under

Alternative Electoral Systems. American Journal of Political Science, Volume 53 Issue 4.

Nelson, Thomas E, and Donald R Kinder. 1996. Issue Frames and Group-Centrism in

American Public Opinion. Journal of Politics 58 (4):1055–78

Nelson, Thomas E, Zoe M Oxley, and Rosalee A Clawson. 1997. Toward a Psychology of

Framing Effects Political Behavior 19:221– 46

OECD. 2009. Agricultural Policies in OECD Countries: Monitoring and Evaluation. Geneva:

OECD.

Olson, Mancur. 1965. The logic of collective action : public goods and the theory of groups.

Cambridge, Mass.: Harvard University Press.

O’Rourke, Kevin, and Richard Sinnott. 2002. The Determinants of Individual Trade Policy

Preferences In Brookings Trade Forum, edited by Susan M Collins and Dani Rodrik,

157–206 Washington, DC: Brookings Institution

Page, Benjamin I, and Robert Y Shapiro. 1992. The Rational Public: Fifty Years of Trends

in Americans’ Policy Preferences Chicago: University of Chicago Press

42

Park, Jong Hee and Nate Jensen. 2007. Electoral Competition and Agricultural Support for

OECD Countries. American Journal of Political Science. Vol.51, Issue 2.

Rogowski, Ronald. 1989. Commerce and Coalitions Princeton, NJ: Princeton University

Press.

Rogowski, Ronald and Mark Andreas Kayser. 2002. Majoritarian Electoral Systems and

Consumer Power: Price-Level Evidence from the OECD Countries. American Journal of

Political Science, Vol. 46, No. 3 (Jul., 2002), pp. 526-539.

Sasada, Hironori. 2009. Japan’s New Agricultural Trade Policy and Electoral Reform:

‘Agricultural Policy in an Offensive Posture [seme no nousei]. Japanese Journal of

Political Science 9-2.

Schattschneider, E. E. 1935. Politics, Pressures, and the Tariff Englewood Cliffs, NJ:

Prentice-Hall

Scheve, Kenneth F, and Matthew J Slaughter. 2001. Globalization and the Perceptions of

American Workers Washington, DC: Institute for International Economics

Scheufele, Dietram. 2000. Agenda-Setting, Priming, and Framing Revisited: Another

Look at Cognitive Effects of Political Communication, Mass Communication & Society,

2000, 3(2&3): 297–316

Sheingate, D. Adam. 2001. The Rise of the Agricultural Welfare State: Institutions and

Interest Group Power in the United States, France, and Japan. Princeton University Press.

Sniderman, Paul M. 2000. Taking Sides: A Fixed Choice Theory of Political Reasoning In

Elements of Reason: Cognition, Choice, and the Bounds of Rationality, edited by Arthur

Stolper, W. and P. Samuelson. 1941. Protection and Real Wages Review of Economic

Studies 9 (1):58–73

Thies, Cameron G. and Schuyler Porche. 2008. Political Economy of Agricultural

Protection. Journal of Politics. Vol.69, Issue 1.

Thies, Michael. 1998. When will pork leave the farm? Institutional bias in Japan and the

United States. Legislative Studies Quarterly. Vol.23, Issue 4.

Vogel, Steven. 1999. “When Interests Are Not Preferences: The Cautionary Tale of

Japanese Consumers,” Comparative Politics, Vol. 31, No. 2, pp. 187-207

Williamson, John B. 1974 Beliefs about the Motivation of the Poor and Attitudes toward

Poverty Policy. Social Problems 21- 5: 734-747.

43

! ! ! ! ! ! ! ! ! ! ! Tables, Charts and Figures

Figure 1: Consumer Support Estimates (%) for OECD Countries

Source: OECD, Agricultural Policies in OECD Countries: Monitoring and Evaluation 2009. Note: The X-axis is selected OECD countries and the Y-axis is Consumer Support Estimates. Positive value indicates the percentage of consumer expenditures on agricultural products that are subsidized by a government, and negative value indicates the percentage of consumer expenditures on agricultural products that consumers bear to support farmers. Japan’s CSE is indicated by the black bar.

44

Figure 2: Results of the Cabinet Office’s Nationally-Representative

Survey on Food Import, 1987-2008

Note: The X-axis is years and the Y-axis is the percentage of respondents who chose a given answer. The percentage of respondents who chose “domestic production even if it means higher prices” is shown in a black line and % of respondents who chose “foreign import if it means cheaper” is shown in a dotted black line. The rest chose domestic production even if it means higher prices for key commodities such as rice.

45

Group 1 Photos: Producer-Priming (Pictures 1,2, and 3)

Note: The three photos above were used for the producer-priming group. Before showing the photos, we asked: “Please carefully look at the photos below and answer the following questions.” (translated by the author(s)). We asked a follow-up question for each photo: e.g., “Q2: What sector or industry do you think these people work for?” where respondents choose from finance, publishing, law, public sector, and manufacturing.

46

Group 2 Photos: Consumer-Priming (Pictures 4,5, and 6)

Note: The three photos above were used for the consumer-priming. Before showing the photos, we asked: “Please carefully look at the photos below and answer the following questions.” (translated by the author(s)). We asked a follow-up question for each photo: e.g., “Q2: What type of grocery shop is this?” where respondents chose from mom-and-pop shop, super-market, organic food store, department store, or discount store.

47

Figure 3: The Effect of Priming in Aggregate

Proportion of Protectionist Responses (%) for General Trade vs. Food

Import

Note: On the general trade issue, consumer-priming is associated with a 13% point lower proportion of protectionist responses than the control group, and an 18% point lower proportion of protectionist responses than the producer-priming group. On food import, consumer-priming has no systmematic effect, while producer-priming increases the proportion of protectionist responses by 10% point. These differences are significant at the p<0.05 level.

48

Figure 4: Projection Hypothesis: Substantive Impact of “Easiness of

Finding a Comparable Job” on % Protectionist Responses

Note: The X-axis is respondents’ response to a question, “How difficult is it for you to find a similar job with a comparable salary if you quit your job?” The Y-axis is the proportion (%) of protectionist responses (increasing food import is “bad” or “very bad”) for each group. The blue line shows the pattern for the producer-priming group, the red line describes the pattern for the consumer-priming group, and green line is the control group. The graph suggests clear positive correlations between the difficulty of finding a comparable job and the support for agricultural protectionism in the producer-priming group. The consumer and control groups show no such relationship as predicted from the projection hypothesis.

49

Figure 5: Distribution of Responses among the Three Groups by Employment Status

Note: The X-axis is % responses and the Y-axis is respondents’ employment status.

50

Chart 1: Producer and Consumer Interests in the Global Economy

Note: The left chart describes how producer and consumer interests are modeled as a dichotomy and trade-off in the standard trade literature. The right chart describes two sources of complexity in identifying preferences and power of producers and consumers over the issue of food import: (i) overlapping and multifaceted identities as producers & consumers, (ii) consumer interests can lead to either free trade or protectionist preferences.

51

! "#$%&!'(!)**+*,-&!"./#0-!122&3*4!.2!50&&!"0#-&! !

! 6&7#*+8&! 6&,*0#%! 9.4+*+8&!

! :&0;!<#-! <#-! ! =..-! :&0;!=..-!

)%%!>&4?.@-&@*4! AB! 'CB! ADB! 'EB! FB!

G6H''CIJ! ! ! ! ! !

90.-,3&0!K*+L,%+! CB! FDB! AFB! AB! 'B!

G6HMN'J! ! ! ! ! !

O.@4,L&0!K*+L,%+! MB! ''B! AAB! 'AB! DB!

G6HMNIJ! ! ! ! ! !

O.@*0.%!=0.,?! PB! FEB! AMB! NB! 'B!

G6HMNNJ! ! ! ! ! !

! "#$%&!F(!)**+*,-&!"./#0-!122&3*4!.2!5..-!QL?.0*!

! 6&7#*+8&! 6&,*0#%! 9.4+*+8&!

! :&0;!<#-! <#-! ! =..-! :&0;!=..-!

)%%!>&4?.@-&@*4! 'IB! MCB! MNB! PB! 'B!

G6H''N'J! ! ! ! ! !

90.-,3&0!K*+L,%+! 'AB! DDB! MIB! IB! ERF!B!

G6HMNIJ! ! ! ! ! !

O.@4,L&0!K*+L,%+! 'IB! MIB! DEB! CB! FB!

G6HMNAJ! ! ! ! ! !

O.@*0.%!=0.,?! 'IB! MAB! D'B! PB! '!B!

G6HDEEJ! ! ! ! ! !

52

Table 3: Ordered Logit Models of Individual Support for Food Import

Model 1 Model 2 Model 3 (bi)

Models Producer Consumer Producer Consumer Producer Consumer

Sympathy

“Sweat” -0.234 -0.203 -0.230 -0.183 -0.121 -0.233

(0.192) (0.208) (0.192) (0.209) (0.224) (0.231)

Social Network -0.253 0.059 -0.258 0.012 -0.568 0.063

(0.200) (0.203) (0.201) (0.210) (0.241)** (0.230)

Redistribution 0.150 0.063 0.151 0.057 0.193 0.153

(0.077)* (0.082) (0.077)* (0.083) (0.089)** (0.089)*

Farm-to-Table 0.036 -0.065 0.047 -0.085 0.123 -0.056

(0.117) (0.121) (0.122) (0.123) (0.141) (0.136)

Projection

-0.123 0.040 -0.121 0.039 -0.135 0.047

Skill Specificity (0.063)* (0.065) (0.064)* (0.065) (0.074)* (0.071)

Age -0.224 0.035 -0.217 0.012 -0.257 -0.050

(0.084)**

* (0.092) (0.087)** (0.096) (0.103)** (0.107)

Temp -0.343 0.251 -0.350 0.254 -0.543 0.013

(0.258) (0.297) (0.259) (0.297) (0.317)* (0.333)

No Jobs 0.443 0.584 0.436 0.602 0.362 0.571

(0.308) (0.281)** (0.309) (0.282)** (0.353) (0.320)*

Producer

Attributes

Low income 0.020 -0.198 0.023 -0.210 0.165 -0.329

(0.238) (0.248) (0.238) (0.249) (0.279) (0.271)

High income 0.568 -0.031 0.569 -0.045 0.668 -0.203

(0.259)** (0.244) (0.259)** (0.244) (0.303)** (0.273)

NoForeign Transact -0.277 0.011 -0.274 0.005 -0.324 -0.023

(0.195) (0.211) (0.195) (0.211) (0.227) (0.232)

College & beyond 0.080 0.235 0.084 0.253 0.098 0.305

(0.200) (0.226) (0.200) (0.227) (0.235) (0.254)

Consumer

Attributes

Female -0.257 -0.001 -0.260 0.028 -0.069 0.012

(0.243) (0.256) (0.243) (0.259) (0.289) (0.288)

Married 0.367 -0.993 0.360 -0.965 0.644 -1.027

(0.267) (0.290)**

* (0.268) (0.292)*** (0.329)* (0.330)***

Have Kids -0.298 0.942 -0.295 0.932 -0.264 0.944

(0.230) (0.250)**

* (0.230) (0.250)*** (0.276) (0.279)***

Cheap-shopper -0.023 -0.035 -0.024 -0.041 -0.224 -0.059

(0.085) (0.082) (0.085) (0.083) (0.101)** (0.092)

Housewives -0.242 0.371 -0.240 0.363 -0.642 0.598

(0.380) (0.396) (0.380) (0.396) (0.452) (0.454)

Co-op Member -0.248 -0.464 -0.245 -0.464 -0.224 -0.537

(0.227) (0.241)* (0.227) (0.241)* (0.276) (0.272)**

Mobilization

LDP Supporters 0.494 -0.299 0.502 -0.315 0.745 -0.408

(0.294)* (0.308) (0.295)* (0.309) (0.345)** (0.330)

No Party IDs 0.340 -0.297 0.330 -0.275 0.518 -0.503

(0.214) (0.214) (0.216) (0.216) (0.259)** (0.234)**

53

Politically Active n/a n/a -0.026 0.066 -0.097 0.073

(0.075) (0.078) (0.090) (0.087)

Protectionist Info 0.177 -0.212 0.182 -0.201 0.152 -0.237

(0.218) (0.237) (0.218) (0.237) (0.262) (0.258)

Cut_1 -1.938 -1.939 -.016

(.115) (.115) (.099)

Cut_2 .024 .025

(.095) (.095) n/a

Cut_3 2.469 2.471

(.134) (.134) n/a

Cut_4 4.645 4.646

(.313) (.313) n/a

# of Obs 1166 1166 1170

LR chi2 (42)81.86 (44) 82.68 (44) 89.21

Prob > chi2 .0002 .0004 .0001

Note: Control group is used as a base category. Positive coefficients indicate support for food import. Coefficients of all the co-variates are the coefficients of interaction terms between a stimuli and individual attributes.

54

Appendix

Table A-1: The Balanced Panel

Descriptive Statistics of Three Experiment Groups

! Producer Consumer Control

Age 1.89 1.89 1.90

(1.35) (1.33) (1.33)

Female 0.50 0.50 0.49

(0.50) (0.50) (0.50)

Low Income 0.28 0.29 0.26

(0.45) (0.45) (0.44)

High Income 0.18 0.24 0.21

(0.39) (0.43) (0.41)

Skill Specificity 3.86 3.94 3.89

(1.29) (1.24) (1.25)

Protectionist Info 0.253 0.235 0.255

! (0.435) (0.425) (0.436)

Note: Protectionist info is the only variable that could potentially be susceptible to “mediation effects” in experiments (see Imai et al. 2009), but the table above shows that respondents’ self-reported levels of exposure to protectionist discourse does not differ across the experimental groups and thus disconfirming the presence of serious mediation effects.

55

A-2 Descriptive Statistics

Variable Mean St. Dev. Min. Max.

Free trade 2.837 (0.755) 1 5

Food import 2.396 (0.857) 1 5

High!income 0.210 (0.407) 0 1

Low income 0.275 (0.447) 0 1

Age 43.292 (13.233) 20 65

Skill specificity 3.896 (1.261) 1 5

College 0.388 (0.488) 0 1

Fully Employed 0.545 (0.498) 0 1

No Job 0.294 (0.456) 0 1

Temp 0.188 (0.391) 0 1

No foreign transactions 0.646 (0.478) 0 1

Sweat 0.483 (0.500) 0 1

Social Network 0.375 (0.484) 0 1

Redistribution 3.557 (1.017) 1 5

Married 0.643 (0.479) 0 1

Cheap shopper 1.956 (1.206) 0 5

!"#"$! 0.251 (0.433) 0 1

Protecionist_info 0.248 (0.432) 0 1

Farm-to-Table 1.01 (0.868) 0 2

Ldp 0.157 (0.364) 0 1

Dpj 0.203 (0.402) 0 1

No party 0.505 (0.500) 0 1

Politically Active 1.128 1.475 0 6

N=1200 ! ! ! !

56

Table A3 Determinants of Individual Support for Food Import: Control Group (1) (2) (3:Binomial) (4:Binomial) Sympathy Sweat 0.021 0.035 -0.010 -0.025 (0.11) (0.18) (0.05) (0.12) Redistribution 0.084 0.085 0.058 0.056 (0.82) (0.82) (0.54) (0.52) Social Network 0.048 0.051 (0.23) (0.22) Farm-to-table 0.020 0.027 0.016 0.015 (0.17) (0.23) (0.13) (0.12) Projection SkillSpecificity -0.057 -0.047 -0.008 -0.016 (0.67) (0.55) (0.08) (0.17) Age -0.020 -0.021 -0.092 -0.084 (0.24) (0.23) (0.93) (0.84) Temp -0.023 -0.028 0.000 0.007 (0.08) (0.09) (0.00) (0.02) Producer Attributes High Income -0.030 -0.049 -0.321 -0.292 (0.11) (0.18) (1.11) (1.01) Low Income 0.026 0.002 -0.066 -0.037 (0.11) (0.01) (0.24) (0.13) College -0.067 -0.083 -0.153 -0.133 (0.32) (0.40) (0.67) (0.58) No job 0.612 0.597 0.526 0.542 (2.24)** (2.18)** (1.72)* (1.77)* No foreigntrans -0.293 -0.271 -0.395 -0.420 (1.38) (1.28) (1.71)* (1.81)* Female -0.031 -0.053 -0.144 -0.119 (0.12) (0.20) (0.50) (0.42) Consumer Attributes Married -0.362 -0.351 -0.216 -0.224 (1.34) (1.30) (0.73) (0.75) Have Kids 0.436 0.420 0.281 0.294 (1.81)* (1.75)* (1.07) (1.11) Cheap Shopper 0.061 0.072 0.042 0.030 (0.69) (0.82) (0.43) (0.31) Housewives -0.488 -0.501 -0.558 -0.561 (1.23) (1.26) (1.26) (1.26) Co-op -0.401 -0.399 -0.385 -0.373 (1.61) (1.58) (1.40) (1.36) Political Mobilization Protection_info -0.030 -0.030 -0.051 -0.056 (0.13) (0.14) (0.21) (0.23) LDP 0.022 0.113 0.116 (0.07) (0.35) (0.35) No Party -0.044 -0.194 -0.152 (0.20) (0.77) (0.60) PoliticallyActive -0.023 -0.081 -0.078 (0.33) (1.02) (0.97) Observations 375 378 378 375 Absolute value of z-stats in parentheses; *significant at 10%;** 5%; *** 1%

57

A4 Individual Support for Food Import: Nationally-representative GLOPE Survey (1) (2) (3:binomial) (4:binomial) Sympathy Inequality -0.144 -0.156 0.051 0.045 (2.27)** (2.47)** (0.55) (0.49) Projection SkillSpecific -0.070 -0.096 -0.143 -0.162*

(0.90) (1.25) (1.44) (1.65) Temp 0.234 0.306 0.560 0.617 (1.18) (1.56) (1.89)* (2.09)**

No job -0.109 -0.164 -0.325 -0.382 (0.28) (0.42) (0.64) (0.76) Age -0.034 -0.075 -0.107 -0.104 (0.62) (1.47) (1.31) (1.38) Producer Attributes College -0.029 0.032 0.044 0.101 (0.19) (0.20) (0.20) (0.45) High Income 0.308 0.292 0.045 0.037 (1.85)* (1.76)* (0.17) (0.14) Low Income -0.116 -0.134 -0.131 -0.123 (0.76) (0.88) (0.55) (0.52) Female -0.568 -0.506 -1.248 -1.237

(4.05)*** (3.65)*** (5.31)*** (5.29)***

Consumer Attributes Coop -0.057 -0.121 0.395 0.372 (0.33) (0.71) (1.48) (1.40) Ldp 0.375 0.235 (2.45)* * (1.02) No Party 0.234 -0.047 (1.52) (0.20) PoliticAct -0.046 0.011 (1.22) (0.20) Agr Coop -0.669 -0.655 (3.49)*** (1.89)*

Constant/Cut1 -3.728 -3.941 -0.814 -0.744

(0.461) (0.446) (1.37) (1.31) Cut Point 2 -1.482 -1.729 n/a n/a (0.444) (0.427) Cut Point 3 0.865 0.583 (0.443) (0.424) Cut Point 4 3.119 2.831 (0.498) (0.481) Observations 935 935 942 942 Absolute value of z-statistics in parentheses * significant at 10%;** significant at 5%; *** significant at 1%

58

A-5: Cut Points for Table A3

Cut_1 -1.868 -1.886 -0.559 -0.531 (0.590) (0.629) (0.672) (0.675)

Cut_2 -0.0206 -0.046

(0.579) (0.619)

Cut_3 2.506 2.507

(0.602) (0.619)

Cut_4 4.536 4.535

(0.760) (0.790)