Menu design approaches to promote sustainable vegetarian food choices when dining out

https://doi.org/10.1016/j.jenvp.2021.101721Get rights and content

Highlights

  • Promoting plant-based diets is needed to reduce greenhouse gas emissions.

  • The food sector can reduce its carbon footprint via scalable menu design approaches.

  • Menu composition must be 75% vegetarian to encourage diners to abandon meat.

  • Vegetarian V-symbols are not used as choice exclusion heuristics by meat eaters.

Abstract

Shifting dietary choices towards vegetarian food is an urgent challenge given the environmental impact of livestock production and imminent need to reduce global greenhouse gas (GHG) emissions. Previous research has proven the value of low cost, scalable menu design interventions to influence food choices, without the need for large-scale educational campaigns. Here, we present two online randomized control trials to determine the effectiveness of two menu design approaches to nudge participants' food choices away from meat and towards vegetarian dishes. In study one we explore the impact of vegetarian items availability on choice. Participants were allocated to menus in which 75%, 50% or 25% of items were vegetarian. We show that meat eaters were significantly more likely to choose a vegetarian meal when presented with a menu with 75% vegetarian items, but not when half (50%) were vegetarian. This finding highlights that saturating the choice environment is required to promote vegetarian food. In study two, we explore the impact of vegetarian symbols (V) to determine if these are used by meat eaters as exclusion decision filters, as is seen in previous work with menus containing ‘vegetarian’ dish sections. Here we show that placement of V symbols, to either the left or right of a dish label, has no impact on choice. These studies provide insights into how the environmental footprint of the food service sector can potentially be reduced using easy and scalable menu design approaches.

Introduction

The 2015 Paris Agreement aims to keep global temperature rises below 1.5° to avoid the worse consequences of climate change (UNFCCC, 2020). In ordet to achieve this, countries around the world must rapidly reduce their greenhouse gas (GHG) emissions, especially within sectors that contribute most to national footprints. The food system is one such sector, with estimates showing that production, processing, distribution, preparation and consumption of food accounts for around 25–30% of all global GHG emissions (Poore & Nemeck, 2018; Crippa, Solazzo, Guizzardi et al., 2021; UNEP, 2020).

Moreover, further analyses shows that, even if GHG emissions from all other sectors were immediately curbed, the impact of the food system alone, if unchallenged, would prevent achievement of Paris Agreement targets (Clark et al., 2020). This situation is further compounded by population growth, projected to reach around 10 billion people by 2050. Therefore, it is essential to find ways to feed approximately one third more people while simultaneously preventing agricultural expansion into virgin forests and reducing GHG emissions. This will require more efficient means of both producing and consuming food (Willett et al., 2019).

Particularly problematic from the perspective of diet efficiency is over-consumption of meat from ruminant animals (i.e. beef, goats and lamb). Ruminant meat is far more resource intensive to produce than vegetarian food. For example, per unit of edible protein, producing beef emits around 20 times more GHG emissions than non-animal sources such as beans, peas and lentils (Ranganathan, Vennard, Waite, Lipinski, Searchinger et al., 2016).

While a wide range of different and promising approaches to improve the efficiency of ruminant meat production exist, these do not negate the need for a global shift in dietary choice towards eating less meat. For example, the EAT Lancet consortium have recommended that, for optimal individual and planetary health, consumption of animal-products must be capped at 98 g of red meat and 203 g of poultry per person per week (Willett et al., 2019), with plant-based foods constituting the majority of the diet.

Given that meat is integral to many cuisines across the world, a crucial question remains as to how exactly we achieve this move to more vegetarian diets? As many years of research and practice in the health domain indicates, eating habits tend to be hard to change. For example, numerous campaigns have been launched worldwide to tackle overconsumption of calories leading to overweight and obesity (Walls, Peeters, Proietto, & McNeil, 2011), yet prevalence continues to rise (Malik, Willet, & Hu, 2020).

Efforts to change dietary choices have tended to focus on educating individuals about associated risks, often via population campaigns or targeted programs directed at ‘high risk’ groups (Stead et al., 2019). Recently, however, research has been directed towards the role of the decision context on food choices (Abrahamse, 2020; Wansink & Love, 2014). More commonly known as ‘nudging’, these interventions involve modifying the way in which a choice is presented, known as the ‘choice architecture’, in dining establishments or food retail. Promisingly, these approaches have shown some efficacy at changing food choices (Vandenbroele, Vermeir, Geuens, Slabbinck, & Van Kerckhove, 2020), often without consumer awareness that their decisions have been influenced (Rust et al., 2020). Examples include modifying the default food offering (i.e. Campbell Arvai, Arvai, & Kalof, 2014), limiting access to the sale of certain food items, redesigning menus (i.e. Feldman, Su, Mahadevan, Brusca, & Hartwell, 2014) , labelling products with symbols, signs or language (i.e Wansink, Painter, & Van Ittersum, 2001), and altering the placement of food products (i.e Dayan & Bar-Hillel, 2011).

Considering nudges to promote sustainable dietary choices, one intervention that has been shown to work well in real-life dining contexts is increasing the availability of vegetarian foods. Examples include adding more vegetarian dishes to menus or buffets, or presenting vegetarian dishes in ways to appear more numerous or abundant (i.e. separating salad ingredients into multiple separate bowls rather than mixing them together in one) (Friis et al., 2017).

Recent research conducted in a university canteen in the United Kingdom showed that doubling the number of vegetarian items on sale (from one to two items on a four-item menu) led to a 62% (range of 40.8%–78.8%) increase in the number of diners choosing these options (Garnett, Balmford, Sandbrook, Pilling, & Marteau, 2019). A second study in a restaurant based in the Netherlands demonstrated that replacing three meat dishes with vegetarian alternatives, and modifying the portion size of meat in a fourth dish, led to a 113% increase in the amount of vegetables consumed and 4% reduction in amount of meat consumed (Reinders, Lieshout, Pot, Neufinger et al., 2020). Similarly, an older campus-based restaurant study found that offering diners a default vegetarian menu, with meat available on a separate menu displayed 3.5m away, significantly increased the probability that vegetarian meals were chosen compared to when diners received a regular menu (Campbell Arvai et al., 2014).

This research is promising as it shows that preferences can be influenced via relatively minor modifications to the way in which a choice is presented, without the need to educate or consciously persuade individuals to alter their behavior. However, despite these initial positive findings, research is yet to determine exactly how much meat availability needs to decrease in order for this approach to produce the desired effect. For example, in the university canteen study noted above, meat options were decreased by 33% (Garnett et al., 2019), while in the Netherlands restaurant study, three meat dishes were replaced by vegetarian dishes in a buffet, although we are not told what proportion of the total dishes on offer this represents (Reinders, Lieshout, Pot, Neufinger et al., 2020).

This question has important practical significance, as knowing exactly how much meat to remove from menus would give useful, pragmatic guidance for retailers and food service operators. For example, to what extent do menus need to substitute meat to vegetarian dishes? Given that diners tend to consume more meat when eating out (Horgan, Scalco, Craig, Whybrow, & Macdiarmid, 2019), reducing the number of meat options may have important implications for profitability or customer retention, and hence, may limit operator willingness to adopt this approach. As such, it would be useful for restaurants to have guidance on the minimum viable reduction in meat availability required to elicit a significant shift in consumer choice towards more sustainable vegetarian options.

A similarly pragmatic question that remains unanswered regards labelling of vegetarian options. To date, these have tended to be indicated by ‘vegan’ or ‘vegetarian’ labels on menus, signs or packaging. This approach is beginning to also receive interest for its potential to influence consumer food choices (Vlaeminck, Jiang, & Vranken, 2014; Tobi et al., 2019).

One intriguing finding in existing literature is the fact that overtly indicating options are ‘vegetarian’ or ‘free from meat’ seems to reduce the numbers of diners willing to purchase them (Bacon & Krpan, 2018; Krpan & Houtsma, 2020). For example, an online menu study found that separating vegetarian items into their own dedicated and labelled ‘vegetarian’ section more than halved the odds they were chosen (Bacon & Krpan, 2018). This finding was replicated in a separate online study also demonstrating that a designated ‘vegetarian’ menu section (versus an environmental or social designation, or no designation) was the least effective approach to promoting vegetarian dishes (Krpan & Houtsma, 2020). In this study, authors suggested that the apparent choice-inhibiting effect of vegetarian labelling owed to this framing leading consumers to believe that vegetarian dishes were less enjoyable, and were used by meat-eaters as exclusionary criteria when scanning menus. While it should be noted that both studies used hypothetical food choices and not real-world behaviours, Piernas et al. (2021) explored the influence of moving vegetarian products to meat aisles in supermarkets on real purchasing decisions. This study found that integrating these products increased sales of meat-free products, yet did not significantly reduce sales of meat products, although this may not apply to restaurant dining where the total number of products on offer may be more restricted.

The implications of this research presents a quandary for food service providers wishing to label vegetarian options without alienating their customer base; how can meat-free options be highlighted for those consumers who want to easily identify them, and also comply with allergen labelling requirements, whilst not inadvertently discouraging meat-eaters to select more vegetarian options?

Thus far, one approach routinely used in many dining and retail establishments is the inclusion of ‘V’ symbols to denote either vegetarian or vegan options. However, no research has yet been conducted to understand whether symbols also reduce the likelihood of these items being selected by consumers. A recent calorie labelling study measured the influence label placement has on dish choices. The results showed that only calorie labels presented before dish titles, but not after, encouraged less calorific food choices. This result was shown to be the case using participants from both the US (who read left to right) and Israel (who read right to left) (Dallas, Liu, & Ubel, 2019) . This finding has yet to be replicated in the context of climate labelling and such research would provide food operators insight into optimal placement of vegetarian symbols to avoid deterring vegetarian food choices.

In this study, we present findings from two separate online randomized controlled trials exploring the influence of vegetarian food availability and vegetarian labelling on choice. These two distinct areas are united by the direct, practical guidance they offer; the first study will help to provide the food service sector with guidance on how to re-design their menu to encourage vegetarian choices, and the second on how to communicate about these options on menus in a way that will appeal to meat-eaters. In study one, we test a range of menus with different ratios of vegetarian: meat dishes to address the question of how much of a menu needs to be vegetarian to encourage diners to shift away from choosing meat? In study two we test the role that placement of ‘V’ symbols on menus play in influencing food choice, answering the question of whether placement of the symbol before or after dish titles influences the number of diners choosing vegetarian items? The goal of both studies is to give practical guidance on the use of these nudge approaches to food service providers wishing to promote more vegetarian options.

Section snippets

Study design

This online randomized controlled trial was delivered via Qualtrics. Participants were randomly allocated to one of three conditions that presented them with a series of menus with different availability of vegetarian and meat dishes – a meat: vegetarian ratio of (a) 2:6 (75% vegetarian menu); (b) 4:4 (50% vegetarian menu); and (c) 6:2 (25% vegetarian menu). Within each condition participants viewed five mock menus that reflected the types of offerings available in popular restaurant chains in

Study design

Study two also used an online randomized controlled design delivered via Qualtrics. Here, participants were randomly allocated to one of three conditions; where menus were presented with the vegetarian symbol (V) to (a) the left (V Left), or (b) the right (V Right) of the dish name, or (c) a control condition where vegetarian items were not highlighted by V symbols. Within each condition, participants viewed a total of eight mock menus, the five included in study one (e.g. a burger menu, a

Summary of findings

This paper presents findings from two online studies that explored whether easy and scalable menu re-design approaches could shift food choices towards more sustainable vegetarian options. In study one, we examined whether the ratio of meat to vegetarian options on menus influenced participants’ choices by comparing a predominantly meat-based menu (25% vegetarian dishes), a predominantly vegetarian menu (75% vegetarian dishes) and a menu offering parity between meat and vegetarian choices. Our

Conclusion

Here we explore the efficacy of two menu-based nudges intended to promote vegetarian food. Our results indicate that availability of vegetarian food is a key factor when presenting options to diners in the hope of promoting more sustainable choices. We demonstrate that predominantly vegetarian menus (>75%) can lead to more vegetarian choices in meat eaters, while still offering a small range of meat options. Furthermore, we show that unlike segregated vegetarian dish sections, vegetarian symbol

Declaration of competing interest

The authors report no competing interests.

Acknowledgements

We would like to thank the participants who took part in this study.

References (38)

  • W. Willett et al.

    Food in the anthropocene: The EAT-lancet commission on healthy diets from sustainable food systems

    Lancet (London, England)

    (2019)
  • W. Abrahamse

    How to effectively encourage sustainable food choices: A mini-review of available evidence

    Frontiers in Psychology

    (2020)
  • L. Bacon et al.

    The language of sustainable diets: A field study exploring the impact of renaming vegetarian dishes on U.K. Café menus, (december), 1–20

    (2018)
  • V. Campbell Arvai et al.

    Motivating sustainable food choices: The role of nudges, value orientation, and information provision

    Environment and Behavior

    (2014)
  • M.A. Clark et al.

    Global food system emissions could preclude achieving the 1.5° and 2°C climate change targets

    Science

    (2020)
  • M. Crippa et al.

    Food systems are responsible for a third of global anthropogenic GHG emissions

    Nature Food

    (2021)
  • S.K. Dallas et al.

    Don't count calorie labeling out: Calorie counts on the left side of menu items lead to lower calorie food choices

    Journal of Consumer Psychology

    (2019)
  • E. Dayan et al.

    Nudge to nobesity II: Menu positions influence food orders

    Judgment and Decision making

    (2011)
  • C. Feldman et al.

    Menu psychology to encourage healthy menu selections at a New Jersey University

    Journal of Culinary Science & Technology

    (2014)
  • Cited by (21)

    View all citing articles on Scopus
    View full text