The (still) mysterious case of agricultural protectionism

ABSTRACT

Existing research demonstrates why farmers demand subsidies, but remains ambiguous as to why consumers/taxpayers support or rather do not oppose such subsidies. We approach this puzzle from two angles: how sensitive are citizens to cost implications of agricultural subsidies, and what is their “value function” with respect to agricultural policy? We argue that farm subsidies, besides benefiting farmers, promise to generate an array of non-market goods that serve various interests in society and thus receive strong support overall. To test our argument, we conducted conjoint survey experiments in two countries: Switzerland and the United States. Our results show that while cost implications only marginally reduce support for subsidies, support is positively affected by the allocation of subsidies to various policy goals, such as guaranteeing food security and enhancing animal welfare. These findings suggest that individual-level support for agricultural subsidies does not result from a lack of information, but reflect genuine appreciation of the perceived multi-functionality of agricultural subsidies.

Las investigaciones existentes demuestran por qué los agricultores exigen subsidios, pero siguen siendo ambiguas en cuanto a por qué los consumidores/contribuyentes apoyan o más bien no se oponen a dichas subvenciones. Abordamos este enigma desde dos ángulos: ¿qué tan sensibles son los ciudadanos a las implicaciones de los costos de los subsidios agrícolas y cuál es su “función de valor” con respecto a la política agrícola? Sostenemos que los subsidios agrícolas, además de beneficiar a los agricultores, prometen generar una variedad de bienes no comerciales que sirven a varios intereses de la sociedad y, por lo tanto, reciben un fuerte apoyo en general. Para probar nuestro argumento, realizamos experimentos de encuestas conjuntas en dos países: Suiza y Estados Unidos. Nuestros resultados muestran que, si bien las implicaciones de costos solo reducen marginalmente el apoyo a los subsidios, el apoyo se ve afectado positivamente por la asignación de subsidios a varios objetivos políticos, como garantizar la seguridad alimentaria y mejorar el bienestar animal. Estos hallazgos sugieren que el apoyo a nivel individual a los subsidios agrícolas no se debe a la falta de información, sino que refleja una apreciación genuina de la multifuncionalidad percibida de los subsidios agrícolas.

Des recherches existantes démontrent les raisons pour lesquelles les agriculteurs demandent des subventions mais restent ambiguës sur celles pour lesquelles les consommateurs/contribuables soutiennent, ou plutôt ne s’opposent pas à de telles subventions. Nous adoptons une approche de ce casse-tête sous deux angles : dans quelle mesure les citoyens sont-ils sensibles aux coûts impliqués par les subventions agricoles, et quelle est leur « fonction de valeur » pour ce qui est de la politique agricole ? Nous soutenons qu’en plus de profiter aux agriculteurs, les subventions agricoles promettent de générer toute une série de biens non marchands qui serviraient divers intérêts de la société et bénéficient donc globalement d’un solide soutien. Pour mettre notre argument à l’épreuve, nous avons mené des expériences d’enquête conjointes dans deux pays : la Suisse et les États-Unis. Nos résultats montrent que bien que les coûts impliqués ne réduisent que marginalement le soutien des subventions, ce soutien est positivement affecté par l’affectation de subventions à divers objectifs politiques, tels que la garantie de la sécurité alimentaire et l’amélioration du bien-être animal. Ces conclusions suggèrent que le soutien au niveau individuel des subventions agricoles ne résulte pas d’un manque d’information, mais qu’il reflète plutôt une véritable appréciation de la multifonctionnalité perçue des subventions agricoles.

KEYWORDS:

• Agricultural protectionism
• collective action theory
• multi-functionality of agriculture
• conjoint survey experiment
• public opinion

Notes

¹ OECD. 2018. Producer Support Estimate. https://stats.oecd.org/index.aspx?queryid=89908#. The PSE is calculated as the ratio of the average price received by agricultural producers to the border price of the products concerned and include: direct monetary payments to farmers, export subsidies, and tariff barriers that the government imposes on certain imported agricultural products.

² At this point, it is important for us to clarify that while we are interested in explaining the lack of opposition toward agricultural subsidies, our argument is not necessarily confined to agriculture. It might as well hold for other highly subsidized sectors in industrialized countries with perceived multi-functionality. One example that comes to mind is the energy sector. We return to this issue in greater detail in the discussion.

³ Although many actors, such as the media, international organizations, or NGOs, try to provide information on the costs of agricultural subsidies in industrialized countries, observational survey data on public opinion toward agricultural subsidies over time, if it existed, is not sufficient to test our cost argument. This is because without an experimental study design through which we can manipulate individuals’ exposure to information about the costs of such subsidies, we cannot know whether such information actually reaches the majority of the general public and is being understood as such. We therefore implemented an experimental design that allows us to isolate the effect of information about the costs of agricultural subsidies and to test its causal effect on individual preferences toward agricultural protectionism.

⁴ Ipsos drew a representative sample of the American and Swiss voting population aged 18 and 65 years old based on their online-access panel. The sample was drawn on quotas based on age, gender, religion, and education.

⁵ There is one exception: In the Swiss sample, high level of animal protection generates a statistically significant higher level of support for the policy proposal than a medium level of animal protection.

⁶ According to Leeper, Hobolt, and Tilley (2020), it is important to rely on marginal means instead of average marginal component effects because differences in support levels in the subgroups might differ and can make any evaluation relying on relative comparisons futile.

Additional information

Funding

This work was supported by the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung.