Our Thoughts on the Cost-Effectiveness of Retention Programs

Introduction

By reducing their meat consumption, individuals can spare a significant number of animals and prevent years of suffering on factory farms. All of our current top charities have programs designed to encourage people to eat fewer animal products. Some of them also have retention programs that are designed to help current vegetarians and vegans maintain their diets. Examples of retention programs include grassroots programs with local meetup groups, Veg Fests and other public events, informational blogs and newsletters, guides to veg-friendly restaurants, and online forums that provide social support.

Retention programs could plausibly make a big difference for animals by reducing vegetarian recidivism. In this analysis, we estimate the monthly rate of recidivism in order to calculate the maximum impact of a retention program with a given number of participants. We then compare the estimated cost-effectiveness of retention programs to the estimated cost-effectiveness of outreach programs like leafleting or online ads. We conclude that retention programs with randomly selected participants are unlikely to be as cost-effective as outreach such as leafleting or online ads programs, but that retention programs can significantly increase their cost-effectiveness by targeting participants who are especially at risk of recidivism.

The Vegetarian Recidivism Rate

We estimate that approximately 31% of new vegetarians return to eating meat within 3 months, approximately 48% return to eating meat with one year, and approximately 84% of vegetarians return to eating meat within their lifetime. We estimate that the average monthly rate of vegetarian recidivism is about 0.7%. In other words, if we randomly selected 1000 vegetarians, about 7 of them would return to eating meat within one month.

Methodology

Our calculations are based on a survey conducted in 2014 by Faunalytics. We relied on their data about the number of current and former vegetarians (including vegans) and the lengths of time that they abstained from eating meat.

According to Faunalytics, just 1.9% of Americans are current vegetarians and 12.1% are either current or former vegetarians. The number of former vegetarians is about 84% of the total number of current and former vegetarians, so Faunalytics concludes that 84% of vegetarians eventually abandon their diet. We think that a vegetarian’s lifetime risk of recidivism may be even higher than 84% since some current vegetarians will likely abandon their diets. However, we use a conservative estimate of 84% in the calculations described in this blog post.¹

Faunalytics also collected data on what we call “vegetarian lifespans,” the lengths of time former vegetarians adhered to their diets. For example, 34% of former vegetarians reported adhering to their diets for 0–3 months. 8% of former vegetarians reported that they did not know how long they adhered to their diets. Assuming that those who did not know how long they adhered to their diets fell into the same distribution of lifespans as those who remembered, we conclude that about 37% of former vegetarians adhered to their diets for 0–3 months.²

In order to determine the probability that a randomly selected vegetarian will stick to their diet for a given length of time, we multiplied the portion of former vegetarians who abandoned their diets per category by the probability that a randomly selected vegetarian will become a former vegetarian: 0.84. For example, we multiplied 0.37 and 0.84 to arrive at our estimate that 31% of all vegetarians will remain vegetarian for 0–3 months.

To determine the portion of vegetarians who will abandon their diets during an average month within each time period, we divided the portion who will abandon their diets during each time period by the number of months in that time period. For example, we divided .31 by 4 to estimate the portion of vegetarians who will abandon their diets in an average month within the 0–3 month range:

Finally, we took the average monthly rate of recidivism for vegetarians in all time periods weighted by the probability that they are in each time period (provided experimentally by Faunalytics). This calculation yielded our estimate that the average monthly rate of recidivism is about 0.7%.

Note that since we averaged the monthly rates for vegetarians in all time periods, our estimate is meant to be applied to a random sample of vegetarians. As you can see in Table 2, the rate of recidivism among individuals who have been vegetarian for 0–3 months is relatively high and the rate of recidivism among individuals who have been vegetarian for 10 years or more is relatively low.

The Cost-Effectiveness of Retention Programs

Animal advocates who must decide how to allocate limited resources may wonder: how does the cost-effectiveness of a retention program compare to the cost-effectiveness of outreach programs like leafleting or online ads?

According to our best estimate, approximately 0.7% of the individuals who click on online ads become vegetarian. Since it costs approximately $0.25 to get one individual to click on an online ad in the United States, the cost of creating one vegetarian through an online ad is about $35.71.⁵ Since the average monthly rate of recidivism is 0.7%, a retention program that targets randomly selected vegetarians and prevents all of its at-risk participants from recidivism would prevent about 0.7% of its participants from recidivism each month. Such a program could cost $0.25 per participant per month and it would be as cost-effective as an online ads program.⁶

We don’t expect any program to prevent all of its at-risk participants from recidivism, of course, but we do expect the participants in retention programs to be easier to influence than the participants of most outreach programs. After all, participants in retention programs have already made the decision to go veg and are likely sympathetic to the many reasons to stay veg. Simply being a member of a veg group may be enough to prevent some vegetarians from abandoning their diets.⁷ If a retention program prevents just half of its at-risk participants from recidivism, it could spend up to $0.12 per participant per month and it would be as as cost-effective as an online ads program.

We find it unlikely that a retention program could be so successful at such a low cost per participant, so we doubt that most retention programs are as cost-effective as most online ads programs. As we learn more about individual retention programs and individual outreach programs, we may find that the best retention programs are comparably cost-effective to at least some outreach programs.

One way for retention programs to improve their cost-effectiveness is by targeting at-risk groups. For example, by actively recruiting participants who have only been vegetarian for 0–3 months, a retention program with 1000 participants might prevent up to 77 vegetarians from abandoning their diets in a single month. In that case, it could cost up to $2.71⁸ per participant per month and rival the cost-effectiveness of online ads, if it prevented all 76 of its at-risk participants from recidivism. If it prevented just half of its at-risk participants from recidivism, it could cost $1.36 per participant per month and it would be as cost-effective as online ads.

Avenues for Further Research

We hope that future research will allow us to refine our estimate of the vegetarian recidivism rate. If more studies like Faunalytics’ were repeated year after year, we might get a better sense of how many people adopt and abandon vegetarianism over time. We would love to see a large-scale longitudinal study of vegetarian retention.

We also hope that future studies will investigate the recidivism rate of vegetarians who participate in specific retention programs. Perhaps organizations that run retention programs could periodically survey their participants to determine whether the recidivism rate among their participants is any lower than our estimate of the average vegetarian recidivism rate. That way, we could discover how much recidivism these programs actually prevent.

Finally, we hope to learn more about the demographics of the vegetarians who abandon their diets and their reasons for doing so. This information might help retention programs target vegetarians who are at a high risk of recidivism.

1. If the lifetime risk of recidivism is higher than 84%, that wouldn’t significantly affect the rest of our estimates. If the lifetime risk of recidivism is as high as 100%, we would estimate that the average monthly rate of recidivism is 0.77% rather than our current estimate of 0.66%. 

2. Plausibly, the former vegetarians who did not know how long they adhered to their diets were vegetarian for less time, on average, than those who could remember. Even if that’s the case, it wouldn’t affect our estimates very much. 

3. Some lifespans are not represented by any of the categories. For example, a lifespan of 3.5 months is not technically included in the “0–3 months” category or the “4–11 months” category. We expect that a respondent who has been vegetarian for 3.5 months would choose the “0-3 months” option, so we concluded that the number of months in the “0–3 months” category is 4, rather than 3. Similarly, we concluded that the number of months in the “4–11 months” category is 8 because it includes those who have been vegetarian for between 11 and 12 months, and so on. 

4. We are treating the “10+ years” category as “10-60 years,” so the range is 50 years or 600 months. 

5. 1/0.007 x 0.25 = 35.71 

6. This assumes that preventing one vegetarian from receding is about as impactful as creating one new vegetarian. 

7. Haverstock and Forgays (2012) find that current animal product limiters are more likely than former limiters to belong to a veg group, including meetup groups and online message boards, ² (1, N = 247) = 9.97, p < .05. 

8. 35.71 x 0.076 = 2.71