Differences in global risk perceptions of biotechnology and the political economy of the media

Int. J. Global Environmental Issues, Vol. 8, Nos. 1/2, 2008 77

Copyright © 2008 Inderscience Enterprises Ltd.

Differences in global risk perceptions of biotechnology and the political economy of the media

Kynda R. Curtis* Department of Resource Economics, University of Nevada, Reno, Reno, NV 89557, USA Fax: +775 784 1342 E-mail: [email protected] *Corresponding author

Jill J. McCluskey School of Economic Sciences, Washington State University, Pullman, WA 99164 6210, USA Fax: +509 335 1173 E-mail: [email protected]

Johan F.M. Swinnen Department of Economics, Katholieke Universiteit Leuven, Belgium Fax: +32 16 326796 E-mail: [email protected]

Abstract: The application of biotechnology to agriculture is highly controversial between consumers and lawmakers across the globe, especially in the case of Genetically Modified Foods (GMFs). Consumer attitudes toward GMFs are largely negative in the developed nations. However, studies conducted in Lesser Developed Countries (LDCs) find that consumer attitudes toward GMFs are positive in some cases. Using a theoretical model of the political economy of the media, we argue that the increased time cost of obtaining media information or ‘stories’ in LDCs leads to a lower consumption on average of biotechnology information, especially negative information than that of the developed countries. In addition, ideological influences on media firms in many LDCs leads to differing supplies of biotechnology stories between LDCs and developed nations. Hence, reduced consumption of biotechnology media stories and the potential for increased positive stories contributes to lower risk perceptions among consumers in LDCs.

Keywords: biotechnology; Genetically Modified Foods; GMFs; Lesser Developed Countries; LDCs; media; risk perceptions.

Reference to this paper should be made as follows: Curtis, K.R., McCluskey, J.J. and Swinnen, J.F.M. (2008) ‘Differences in global risk perceptions of biotechnology and the political economy of the media’, Int. J. Global Environmental Issues, Vol. 8, Nos. 1/2, pp.77–89.

78 K.R. Curtis, J.J. McCluskey and J.F.M. Swinnen

Biographical notes: Kynda R. Curtis is an Assistant Professor and State Extension Specialist at the University of Nevada, Reno. She teaches agricultural and food marketing classes for the Department of Resource Economics and leads educational programming in agribusiness for agricultural producer in Nevada. Her research focuses on consumer perceptions of food quality and food labelling. She has published several peer-reviewed journal articles, including in the Canadian Journal of Agricultural Economics and the China Economic Review. She chairs the Western Extension Marketing Committee and WERA-101. She holds a PhD in Economics from the Washington State University (2003).

Jill J. McCluskey is a Professor in the School of Economic Sciences. She teaches graduate-level industrial organisation and is the Chair of Graduate Studies. Her research focuses on food and environmental quality and consumer perceptions. She has published over 40 peer-reviewed journal articles, including in the Review of Economics and Statistics, Journal of Environmental Economics and Management and the American Journal of Agricultural Economics. She is an Associate Editor of the American Journal of Agricultural Economics and the Journal of Industrial Organisation. She received her PhD in 1998 from the University of California, Berkeley.

Johan Swinnen is a Professor of Development Economics and Director of the LICOS Centre for Institutions and Economic Performance at the KU Leuven. His research focuses on institutional reform and development, globalisation, media economics, and agriculture and food policy. His latest books are ‘Global Supply Chains, Standards and the Poor’ and ‘From Marx and Mao to the Market’. He is an Associate Editor of the American Journal of Agricultural Economics and a Coordinator of the European Network of Agricultural and Rural Policy Research Institutes (ENARPRI) and a Member of the Advisory Committee of the Regoverning Markets Global Project. He holds a PhD from Cornell University.

1 Introduction

The application of biotechnology to agriculture is highly controversial between consumers and policy makers across the globe, including in the developing world (Pinstrup-Andersen and Schiøler, 2001). Despite claims from scientists and official institutions that Genetically Modified Food (GMF) products are safe, consumer attitudes toward GMFs are largely negative in many developed countries, especially in the European Union (EU) and Japan (see Moon and Balasubramanian, 2001; Angulo and Gil, 2003; McCluskey et al., 2003; Grimsrud et al., 2004). The US is one exception to this rule, where consumers are largely ambivalent about GMFs (see Nelson, 2001; Chern et al., 2002; Wolf, 2003). However, studies conducted in Lesser Developed Countries (LDCs) find that consumer attitudes toward GMFs are less negative and in many cases positive (see Subrahmanyan and Cheng, 2000; Curtis, McCluskey and Wahl, 2002; Li et al., 2002; Pachico and Wolf, 2002).

A rationale for these differences in attitudes is the diversity of structural economic environments which contribute to the differences in perceived benefits and risks. Consumers in the developed world have much less to gain from biotech-induced farm productivity improvements compared to developing country consumers who have much

Risk perception differences of biotechnology and political economy 79

to gain from cheaper food. This may provide one explanation of why EU consumers are largely against biotechnology and why consumers from LDCs are in favour (Pinstrup-Andersen and Schiøler, 2001). In addition, developing countries have a higher proportion of rural residents. It is likely that consumers who are associated with agriculture have a better idea of the types of amount of pesticides used on non-GM crops than urban consumers. Findings of McCluskey et al. (2003) support this point as they find that the people associated with agriculture are much more in favour of the GM crops than urban consumers. Hence, it is the differences in perceived benefits that drive attitudes.

Another reason for the differences across countries, which we explore in this paper, is the different organisation and structure of the media in developed and LDCs. Several studies find that subjective probabilities of risks and benefits associated with GMFs are the functions of the level of consumer trust in government regulators regarding food supply safety, attitudes toward scientific discovery and the influence of media coverage (Caswell, 2000; Nelson, 2001; Loureiro, 2003; Curtis, McCluskey and Wahl, 2004). In fact, media coverage in the form of newspapers, radio, TV and internet ‘stories’, is often the main source of information consumers use to form attitudes regarding GMFs. Hoban and Kendall (1993) find that over 90% of consumers receive information about food and biotechnology primarily through the popular press and television. Frewer, Howard and Shepherd (1998) conclude that television, radio and newspapers, followed by discussion with other people, were the main information channels by which people base their decisions about biotechnology. Thus, the mechanisms of information provision are essential to our understanding of consumer choice, such as consumer attitudes toward GMFs or lawmaker response to the advances in biotechnology.

The industrial organisation of the media and the demand for information through media differs fundamentally between developed and LDCs. In this paper, we explain how these fundamental differences may have an important impact on information dissemination on GMFs and consequently on people’s attitudes on this issue. More specifically, based on the model of McCluskey and Swinnen (2004), first, we argue that commercial media is more likely to highlight potential risks associated with biotechnology in its reporting – a finding supported by, e.g. Marks and Kalaitzandonakes (2001) who find bias in media coverage of agrobiotechnology in the US and the UK.1

Secondly, the increased cost of obtaining and processing media information in developing countries leads to a lower consumption on average of biotechnology ‘stories’ and to a proportionately stronger reduction in negative stories than therefore to a different bias in media coverage than in developed countries. Thirdly, we argue that the different organisation of the media in LDCs, in particular, the more extensive control of government of the media may, under some conditions, lead, on average, to a more positive coverage of biotechnology in the media. The combination of these factors leads to media coverage, which is less likely to be negatively biased in developing countries and contributes to more favourable perceptions of GMFs and biotechnology among consumers in these countries. This has important implications, since risk perceptions are the main driver of a consumer’s willingness to enter the market for GMFs (Curtis and Moeltner, 2006).

The paper is organised as follows. First, we link media coverage with consumer risk and demand. Then, we present a political economic model of the media. Based on this model, we discuss the implications for perceived risk of biotechnology from the structural differences in the media market between developed and LDCs. Finally, we offer conclusions.

80 K.R. Curtis, J.J. McCluskey and J.F.M. Swinnen

2 Effect of media coverage on risk perceptions and demand

Extensive media coverage can contribute to heightened perceptions of risk, resulting in decreased demand for GM food products. For example, the media coverage of Bovine Spongiform Encephalopathy (BSE), commonly known as ‘mad-cow disease’, in the EU, Japan and Canada, resulted in declining beef demand. The December 2003 discovery of a BSE infected Holstein (dairy) cow in south-central Washington State led to the closure of several overseas markets for the US beef. Verbeke, Ward and Viaene (2000) find that television coverage on meat safety had a negative effect on demand for red meat after the BSE-outbreak in Belgium. Johnson (1988) shows how media coverage of products contamination by the pesticide Ethylene Dibromide (EDB) resulted in important disruptions in the market for grain products.

Negative media coverage of anti-GM groups’ protests over GM foods in the US, forced companies such as McDonalds, Wendy’s and Frito Lay, to stop using GM potatoes in their French fry and potato chip products for fear that consumers would consider the potatoes unsafe and lose confidence in their products. As processed potatoes are a high profit centre for these firms, they are naturally reluctant to take on any unnecessary risks. In addition, public pressure and the threat of major negative media coverage forced Starbucks to use only milk free of recombinant Bovine Growth Hormone (rBGH), which is used to increase milk production in cows, in its coffee drinks.

3 A general model of the media and the information market

Following McCluskey and Swinnen (2004), we present a model which shows that the number of media stories, as well as media story attributes, differ for media firms with non-identical ideological (policy) preferences. Let us first consider the supply side of the market, i.e. the ownership and organisation of the media. We assume that the media organisation is driven by its profit goals, as well as its ideology – reflected in its own story attribute preferences. The attributes (which we will represent as ) include ideology, attitude (e.g. negative vs. positive), format, vocabulary and coverage choice (i.e. regional, local or national). Media revenues result primarily from advertising dollars, a function of constituent size and demographics. Each media firm j, where j = {LDC, Developed} has the following objective function for producing a story in -attribute space,

2( , ) ( ) ( )m m mj j j j j j j jU (1)

where ( j) is the profit function for the media firm. Profits are defined as ( j) = pmj( j) – Fj, where p is the exogenous price of stories and F is the fixed cost of

production. The number of stories on a specific issue, such as biotechnology, is defined

as mj ( j) where ( j) is the set of story attributes. Further, mj is the media firm’s preferred

location in attribute space, which reflects its ideological preferences. The term 2( )m

j j j in the media firm’s objective function reflects that if the media firm

prints stories which deviate from its ideological preferences then it recieves disutility from doing so. Squaring this term makes deviations on either side of the spectrum a

Risk perception differences of biotechnology and political economy 81

negative matter. For example, for a political centrist, stories that are either too far to the right or the left are undesirable. The parameters j and j reflect the relative importance of attribute preferences and profit in the firm’s objective function. If propaganda is all important, as it was for Pravda, the Former Soviet News Agency, then would be large and would be zero. On the other hand, for media tycoon Rupert Murdoch, is very small and is large. McGuire (2001) sums Murdoch’s profit maximising objective function up with,

“Opportunism, not ideology, drives Murdoch. Whether backing Tory or Labour, cold warriors like Margaret Thatcher or communists in Beijing, one aim remains: the desire for a friendly market for his expanding media empire.” (p.21)

The firm chooses its optimal attribute location *j to maximise its objective function,

2max ( ) ( )j

mj j j j j jpm F (2)

resulting in the following first-order condition,

* *2 ( ) ( ) 0.mj j j j jpm (3)

The first-order condition exhibits the firm’s trade-off between ideology and profits. The

firm’s story supply is determined by the ideological choice, *j , which solves the firm’s

first-order condition. Now, let us consider the demand side of the media market. The consumer uses the

biotechnology information provided by the media outlets to form predictions of the probability of the occurrence of risks and benefits from GMF consumption. We assume that the consumption of the media stories will positively affect the consumer’s income through an increase in the reader’s knowledge level.2 The argument behind this assumption is that the reader uses media stories to improve his or her estimate of matters that will affect his or her income such as the financial health of a firm or which politician will be elected. The consumer further obtains positive utility from consuming leisure and all other goods and obtains negative utility from consuming media stories which diverge from his or her attribute preferences.

Consumer i’s utility in countries 1 and 2, I = {1, 2} derived from consuming a media story from outlet j at time t will depend upon a vector of story characteristics,

2( , , , ) ( , ) ( )itc c l l c

j i it it it it it j iU L x g L x m (4)

where litL denotes time spent on leisure for the ith person at time t, xit denotes the

composite good, ci denotes consumer i’s K-dimensional ideal point, j denotes the

K-dimensional location of outlet j issued at time t, mit is the number of stories purchased, weights their ideology preference, and g(.) is an increasing, concave function in both

leisure and the composite good. The squared difference between the location of the media products characteristics and the consumer’s preferred type of media negatively affects utility. That is, consumers receive disutility from reading stories that diverge from their preferred types. We assume that all consumers are identical by country, with the exception of ideological preferences. We assume that there are n consumers whose

82 K.R. Curtis, J.J. McCluskey and J.F.M. Swinnen

preferred ideological locations are uniformly distributed in ideology space. The consumer maximises utility subject to a budget constraint

2

1

; ,1 1

wit it it

it

x m Lf

r r m(5)

and a time constraint,

1 ,l wit it d itL L c m (6)

where r is a discount rate. The price of all the other goods x and the price of mit, the number of stories purchased is normalised to equal one, f(.) is a concave production function, w

itL is the time spent in working and 2 is the variance of expected events that affect the consumer’s welfare. This term, ( 2/mit–1), has a negative effect on income.

An important outcome of our model is the so-called ‘bad news hypothesis’, i.e. the fact that commercial media have the tendency to report disproportionately about negative implications of new developments, including new technologies (see McCluskey and Swinnen (2004) for a formal derivation). This result is driven by the demand side of the market due to media consumers’ preferences, reflecting differential expected welfare effects of negative and positive news.

4 Implications of structural differences in the media market between developed and lesser-developed countries

There are structural differences in the media market both on the demand and the supply side. We start first with the demand side effects, and then discuss supply side factors.

4.1 Costs of media access

Obtaining access to media outlets such as newspapers, radio, television and the internet is much more difficult in LDCs than in developed nations, for two reasons. First, as Table 1 illustrates, the number of information outlets per 1,000 people in selected LDCs is much lower than in selected developed nations. In fact, the average number of daily newspapers and radios in developed countries is almost three times that in LDCs. Developed countries have twice the average number of televisions, seven times the number of personal computers and five times the number of citizens with the internet service.

Secondly, Table 2 shows that both men and women work three to four more hours a week on average in LDCs than in developed countries. This increased work time decreases the amount of time available for leisure activities, including consuming media stories on biotechnology. The combination of both factors has an important impact on consumers’ consumption of media information on, e.g. GMFs. Due to the extra effort in terms of time, the average consumer in LDCs must take in order to reach media outlets, the time cost of ‘consuming’ media stories is much greater than the time cost in developed countries. Formally, this implication follows from condition (5). Notice that 2

is divided by the consumption of media stories. As the consumer increases his or her consumption of media stories, the effect of the variance of events on income diminishes. cd is the average time of ‘consuming’ a story in country d. With the average time of

Risk perception differences of biotechnology and political economy 83

consuming a story in a lesser developed country (1) greater than the average time of consuming a story in country 2 (developed), or c1 > c2 , the time constraint for country 1 is more restrictive than that of country 2.

Table 1 Information outlets per 1,000 people

Country Daily newspapers

(2000) Radios (2001) Television

(2002)

Personal computers

(2002) Internet service

(2002)

Developed

The USA 231 2,117 938 658.9 551

Canada 159 1,047 691 487 513

Japan 578 956 785 382.2 449

Netherlands 306 980 648 466.6 506

Germany 305 570 661 431.3 412

France 201 950 632 347.1 314

Average 296.7 1103.3 725.83 462.18 457.5

Less developed

Mexico 94 330 282 82 98

Peru 0 269 172 43 93

Thailand 64 235 300 39.8 78

India 60 120 83 7.2 16

Chile 98 759 523 119.9 238

China – 339 350 75 27.6

Romania 300 358 697 69.2 83

Bulgaria 116 543 453 51.9 81

Average 104.57 369.13 357.5 61 89.33

Source: World Bank (2004).

In equilibrium, the first-order conditions show that the net marginal benefit of time used in processing media stories is greater for country 2,

22

11

22

22

1( ) (1 )

1( ) (1 )

cj x m x

cj x m x

r g f gc m

r g f gc m

(7)

which results in a higher demand for news stories in country 2, a function of story attributes, consumer preferred attributes and other goods.

84 K.R. Curtis, J.J. McCluskey and J.F.M. Swinnen

2

1 21 1

2

2t 222 2

(1 ) ( , , )

( )

(1 )( , , ).

( )

l

l

c m x1t j c

x jL

cm xjc

x jL

f g rm x

c g g

f g rm x

c g g

(8)

That is, country 1 will consume fewer media stories.

Table 2 Average weekly work hours for men and women, 2000–2003

Country 2000 2001 2002 2003

Developed

The US 41.00 40.60 40.5 42.6

Canada 31.6 31.6 31.9 –

Japan 42.70 42.20 42.20 –

Germany 44.14 43.64 44.41 44.84

France 35.61 35.77 35.24 36.02

The UK 39.8 39.8 39.6 39.6

Average 39.14 38.93 38.97 40.76

Less developed

Mexico 43.8 43.4 44.1 43.2

Hong Kong 44.6 46.6 46.9 46.7

Taiwan 43.25 41.27 41.27 –

Thailand 50.1 – – –

India 47 46 – –

Chile 44.1 43.7 43.8 43.5

Costa Rica 46.60 47.00 47.00 47.00

Romania 37.9 38.3 39.6 39.4

Slovakia 33.2 33.2 32.05 31

Czech Republic 40.7 40.7 40.7 –

Average 43.12 42.24 41.92 41.8

Source: International Labor Organization (ILO) (2004).

Average: Average based on the countries in the table.

Given this and following the ‘bad news hypothesis’, media stories regarding biotechnology are mostly negative, then the average consumer’s beliefs regarding the risks and benefits associated with GMFs will differ across the two countries. More specifically, the lesser developed country consumers will be more accepting of biotechnology.

Risk perception differences of biotechnology and political economy 85

4.2 Ideology and government control over media

In many LDCs, there is government control or governmental influence on media outlets, especially on the types of stories or story attributes which are released to the public. The first source of evidence is a measure of freedom of the press. Freedom House (Sussmann and Karlekar, 2002) assigns a numerical index of freedom of the press and rates each country with one of the three designations: ‘free’, ‘partly free’, and ‘not free’. By these rankings, all the developed countries of Western Europe and North America have a ‘free press’, while the press situation in the less developed countries is more mixed and certainly less free on average. In particular, in countries such as China and Colombia, the press is considered not free (See Table 3).

Table 3 Press freedom indicators

Score* Category

Developed

Germany 16 Free

The US 17 Free

Canada 17 Free

Japan 20 Free

France 20 Free

Less developed

Costa Rica 19 Free

Hungary 21 Free

Chile 24 Free

India 38 Partly free

Peru 40 Partly free

Mexico 42 Partly free

Romania 47 Partly free

Colombia 63 Not free

China 82 Not free

Source: Freedom house.*Note: 0–30 equals ‘Free’ Media; 31–60 equals ‘Partly free’ Media; 61–100 ‘Not free’ Media.

Another source of evidence can be found in Table 4, which provides a governance ranking for selected countries. The percentile rankings indicate the percentage of countries worldwide that rank below the selected country. Categories include ‘voice and accountability’, in which a high ranking is associated with democratic governance and representative responsibility, ‘political stability’, which represents the process by which the authorities are selected and replaced, as well as ‘rule of law’, which represents the respective citizens who have for their governmental institutions. For rule of law, a higher ranking is associated with high citizen respect of the governmental authorities. As is shown, developed nations have much higher rankings on average, indicating a higher occurrence of democratic governance, political stability and citizenry respect for the government officials. In LDCs, there is a higher occurrence of political instability and

86 K.R. Curtis, J.J. McCluskey and J.F.M. Swinnen

non-representative government, which is generally not well respected and/or supported by the citizenry.

Table 4 Governance rankings by percentile (0–100), 2002

Country Income category Voice and

accountability Political stability Rule of law

Developed

The US High 90.9 89.7 (2000) 91.8

Canada High 94.9 87 93.8

Japan High 79.3 90.3 88.7

Germany High 95.5 88.1 92.8

France High 88.4 70.8 87.6

Average 89.9 85.18 90.94

Less developed

Mexico Upper middle 59.6 50.8 52.1

Peru Lower middle 57.6 25.4 40.7

Costa Rica Upper middle 84.8 56.5 72.2

India Low 60.6 22.2 57.7

Colombia Lower middle 38.8 4.9 26.8

Chile Upper middle 84.3 85.9 87.1

China Low middle 10.1 51.4 51.5

Romania Lower middle 61.1 58.4 54.1

Hungary Upper middle 85.4 88.6 78.9

Average 60.26 49.34 57.9

Source: World Bank (2004).

( ): Data year. Average: Average based on the countries in the table.

From these different sources, in general, we argue that on average, governments have more control over media in LDCs. As a result, media outlets in LDCs are more likely to be used by the governmental authorities to control or persuade the public to take certain views on policy. It is easy to see from condition (3) that the ideological bias of the media will be stronger in countries where the government controls the media. With more government control in country 1 (LDC), the importance of ideology in the media supply of stories is larger than that in country 2 (Developed), or 1 > 2. The supply of stories will differ between the two firms, as the quantity supplied is a function of ideological choice and exogenous price.

* * * *1 1 1 1 2 2 2 2( , , , , ) ( , , , , ).m m

j jm p m p (9)

With more government control over the media, the ideology of the government with respect to biotechnology is essential to understand how this factor will affect media coverage. Governments under pressure from the EU, such as in some parts of Africa may use the media to produce anti-GM information. However, in other cases, where positive biotechnology stories are the ideological choice of the government controlled media firm,

Risk perception differences of biotechnology and political economy 87

the supply of positive biotechnology stories will be greater in the LDC media firm than in the developed nation media firm and the supply of negative stories will be lower. China is an interesting case to illustrate this point. The Chinese government has had a strong commitment to biotech research since the early 1990s (Gale et al., 2000). China is one of the largest producers of GM crops in the world with large annual research and development expenditures to support it. As of 2001, China had over 200 biotech companies and over 300 institutions engaged in biotech research and development. The total number of people working in biotech research and development exceeded 5,00,000 (China Commercial Brief, 2001). The government has an important pro-GM preference and has a very strong influence on the media, which is considered the least free in the ranking of Table 4. Therefore, it is not surprising that government controlled media coverage in China concerning GM crops has been positive and has contributed to public support for research and development spending and boost consumer demand for such products. Interestingly, a study by Li et al. (2002) finds that only 9.3% of the survey respondents had somewhat a negative or very negative opinion concerning the use of biotechnology in foods and only 7.8% associated high risks with GMFs.

5 Conclusions

Understanding the political economy of the media is important when considering what factors drive consumer’s risk perceptions regarding GMFs. In this paper, we recognise that the differences in potential benefits and risks of GMFs and differences in trust in state institutions may affect differences in consumer attitudes. In LDCs, concerns of food availability and nutritional intake are much greater in comparison to the US, the EU and Japan. Increased crop yields and dietary supplements provided by GMFs are of greater benefit to LDCs.

In this paper, we explore how differences in media organisation and demand between developed and LDCs can also contribute to the differences in consumer attitudes towards GMFs. Risk attitudes toward new technology are often influenced by the media coverage. Consumers in LDCs have less access to media coverage and governments often exert more control over media in LDCs. These factors have contributed to generally more positive attitudes toward GMFs in developing nations.

We use a political economy model of the media to show that lower perceived risks regarding GMFs from a media influence standpoint can be attributed to the differences in media consumption in LDCs. We show that developing countries are likely to differ from developed countries in three important factors affecting media coverage of GM foods. First, access to media outlets such as newspapers, radio, television and the internet is more difficult in LDCs compared to developed nations. Secondly, people have less leisure time in LDCs. Both factors increase the costs of consuming a media story. Thirdly, in many LDC countries, there is a government control or governmental influence on media outlets, especially on the types of stories or story attributes which are released to the public. In countries such as China, where the government is pro-biotechnology and influences the media, this contributes to pro-GM attitudes in consumers.

88 K.R. Curtis, J.J. McCluskey and J.F.M. Swinnen

Acknowledgements

The authors gratefully acknowledge financial support from the IMPACT Centre at Washington State University and the Nevada Agricultural Experiment Station (Publication No. 51043034).

References

Angulo, A.M. and Gil, J.M. (2003) ‘Consumers’ attitudes and buying intentions towards GM food in Spain’, Selected Paper of the 7th ICABR International conference on Public Good and Public Policy for Agricultural Biotechnology, June 29–July 4, Ravello, Italy.

Caswell, J.A. (2000) ‘An evaluation of risk analysis as applied to agricultural biotechnology with a case study of GMO labeling’, Agribusiness, Vol. 16, pp.115–123.

Chern, W.S., Rickertsen, K., Tsuboi, N. and Fu, T-T. (2002) ‘Consumer acceptance and willingness to pay for genetically modified vegetable oil and salmon: a multiple-country assessment’, AgBioForum, Vol. 5, pp.105–112.

China Commercial Brief (October 26, 2001) Foreign Commercial Service, American Embassy, Beijing, Vol. 2.

Curtis, K.R., McCluskey, J.J. and Wahl, T.I. (2002) ‘Is China the market for genetically modified potatoes?’, AgBioForum, Vol. 5, pp.175–178.

Curtis, K.R., McCluskey, J.J. and Wahl, T.I. (2004) ‘Consumer acceptance of genetically modified food products in the developing world’, AgBioForum, Vol. 7, pp.69–74.

Curtis, K.R. and Moeltner, K. (2006) ‘Genetically modified food market participation and consumer risk perceptions: a cross-country comparison’, Canadian Journal of Agricultural Economics, Vol. 54, pp.289–310.

Frewer, L., Howard, C. and Shepherd, P. (1998) ‘The importance of initial attitudes on responses to communication about genetic engineering in food production’, Agriculture and Human Values, Vol. 15, pp.15–30.

Gale, F., Lin, W., Lomar, B. and Tuan, F. (2002) ‘Is biotechnology in China’s future?’, China’s Food and Agriculture: Issues for the 21st Century. Economic Research Service, USDA, AIB-775, pp.34–37.

Grimsrud, K.M., McCluskey, J.J., Loureiro, M.L. and Wahl, T.I. (2004) ‘Consumer attitudes to genetically modified food in Norway’, Journal of Agricultural Economics, Vol. 55, pp.75–90.

Hoban, T.J. and Kendall, P.A. (1993) ‘Consumer attitudes about food biotechnology’, Raleigh, N.C.: North Carolina Cooperative Extension Service.

International Labor Organization (ILO) (2004) Yearly Labor Statistics. Available at: http://laborsta.ilo.org.

Johnson, F.R. (1988) ‘Economic costs of misinforming about risk: the EDB scare and the media’, Risk Analysis, Vol. 8, pp.261–270.

Li, Q., Curtis, K.R., McCluskey, J.J. and Wahl, T.I. (2002) ‘Consumer attitudes toward genetically modified foods in Beijing China’, AgBioForum, Vol. 5, pp.136–144.

Loureiro, M.L. (2003) ‘GMO food labeling in the EU: tracing ‘the seeds of dispute’, EuroChoices,pp.18–22.

Marks, L.A. and Kalaitzandonakes, N. (2001) ‘Mass media communications about agrobiotechnology’, AgBioForum, Vol. 4, pp.199–208.

McCluskey, J.J., Grimsrud, K., Ouchi, H. and Wahl, T.I. (2003) ‘Consumer response to genetically modified food products in Japan’, Agricultural and Resource Economics Review, Vol. 32, pp.222–231.

Risk perception differences of biotechnology and political economy 89

McCluskey, J.J. and Swinnen, J.F.M. (2004) ‘Political Economy of the Media and Consumer Perceptions of Biotechnology’, American Journal of Agricultural Economics, Vol. 86, pp.1230–1237.

McGuire, S. (2001) ‘Blair vs. The Press’, Newsweek, pp.18–22.

Moon, W. and Balasubramanian, S.K. (2001) ‘Public perceptions and willingness-to-pay a premium for non-GM foods in the US and UK’, AgBioForum, Vol. 4, pp.221–231.

Nelson, C.H. (2001) ‘Risk perception, behavior, and consumer response to genetically modified organisms’, American Behavioral Scientist, Vol. 44, pp.1371–1388.

Pachico, D. and Wolf, M. (2002) ‘Attitudes toward genetically modified food in Columbia’, Presented at the 6th International ICABR Conference in Ravello, Italy.

Pinstrup-Andersen, P. and Schiøler, E. (2001) Seeds of Contention: World Hunger and the Global Controversy over GM Crops. Johns Hopkins University Press, Baltimore.

Subrahmanyan, S. and Cheng, P.S. (2000) ‘Perceptions and Attitudes of Singaporeans toward Genetically Modified Food’, The Journal of Consumer Affairs, Vol. 34, pp.269–290.

Sussmann, L.R. and Karlekar, K.D. (Eds) (2002) The Annual Survey of Press Freedom. New York, NY: Freedom House.

Verbeke, W., Ward, R.W. and Viaene, J. (2000) ‘Probit analysis of fresh meat consumption in Belgium: exploring BSE and television communication impact’, Agribusiness, Vol. 16, pp.215–234.

Wolf, M. (2003) ‘Purchase interest in corn that was grown using biotechnology’, Paper presented in the Proceedings of the 7th International ICABR Conference in Ravello, Italy.

World Bank (2004) World Development Indicators: The Information Age. Available at: http://www.worldbank.org.

Notes 1They find bias in so far as coverage has emphasised different frames (biosafety and food safety) at various points in time depending upon unfolding events.

2The assumption of information affecting consumer utility through income is convenient for our derivations but not necessary for our results.