Sustaining Cooperation in the Local Food Economy Dr. Tim Waring, Mitchell Center Jeremy Bloom, Buying Club Software Taylor Lange, Ecology and Environmental Sciences Afton Hupper, School of Economics

Background 1. Sustainability is not always win-win. 2. Hard sustainability problems are social dilemmas over resource use. 3. Social dilemmas can be solved by cooperation and institutions.

Larger Effort in Evolution of Sustainability National Socio-Environmental Synthesis Center (SESYNC) National Institute for Mathematical and Biological Synthesis (NIMBioS) Jeremy Brooks, Ohio State University Sustainability Science Special Issue (Maine, Tanzania, Bali, Caribbean, California, Canada)

Social Dimension of Local Food ● Local food is growing (USDA ERS, 2015) ● Health, environmental concerns, quality perceptions ● Social value. Local food is socially local, provides social connections. (Tremblay & Waring, 2014)

Evolution of Local Food Organizations ● Cooperation evolves in sustainability dilemmas (Waring et al., 2015). ● Evolution of cooperation and collective action in blueberry industry (Hanes & Waring, 2017), lobster fishery (Waring & Acheson, 2017). ● How does cooperative behavior and supporting institutions emerge and persist in food organizations?

The Case of Food Sovereignty in Maine “Local Food and Community Self-Governing Ordinance” Local Food Rules Heather Retberg Cooperation factor: Municipal Ordinance requires threshold of voter support.

The Food Sovereignty Case ● Democratic votes ● Spread between towns ● Leadership ● Descent with modification ● LD 1648 “Maine Food Sovereignty Act” ● Easy: one-time rule changes (low cooperation)

Cooperation in Co-operatives ● Ethan Tremblay, MA Economics ● Are food co-operatives more cooperative? ● Experimental measurement of unenforced fairness, altruism (Dictator game) Ethan Tremblay MA Economics

The Dictator Game, a measure of cooperation

Cooperation in Cooperative Organizations Blue Hill Food Co-op Blue Hill Tradewinds Grocery

Results Tremblay & Waring forthcoming % difference in contribution ● Tobit Regression model ● Site effect still dominates ● Even accounting for age, income, sex, education, other variables $10 $9 $8 $7 $6 $5 $4 $3 $2 $1 $0 | | | | | | | -15% -10% -5% 0% 5% 10% 15% Dollars Donated More Grocery Donations More Co-op Donations

Cooperative Food Buying Clubs ● Local food entrepreneur ● Portland area leader ● Internet Farmer - Marketing & online ordering for local food. ● BuyingClubSoftware.com - help people work together by buying in bulk to save time and money. Jeremy Bloom Founder, BuyingClubSoftware.com

Splits

Buying Clubs in the Food System PRODUCERS Farmers, Makers, Preservers Vegetables, Meat, Dairy, Jams, Jellies, Maple Syrup, Honey, Kimchi, Soy Sauce, Baked goods DISTRIBUTORS Small Markup Amy’s Soups Toilet Paper GROCER 40-100% Markup Hannaford Walmart Amazon/Whole Foods Trader Joes BUYING CLUB

Community

Is cooperation involved in splits? What is cooperation anyway?

How does Cooperation Evolve? Classic Prisoner's Dilemma Defectors Outcompete Cooperators, So how can there be cooperators? Player A, B Defect Cooperate Defect 1,1 4,0 Cooperate 0,4 3,3

Cooperation Evolves at the Group Level The group of Cooperators has an edge over the groups with all defectors “Selfish individuals outcompete altruistic individuals within groups. Altruistic groups defeat selfish groups. Everything else is commentary.” -Wilson & Wilson, 2007 Group Type Individual Fitnesses Group Fitness (Mean) 2 Defectors 1 , 1 1 1 Defector + 1 Cooperator 4 , 0 2 2 Cooperators 3 , 3 3

How to get to a Group of Cooperators? Engage in reciprocity (Trivers 1971, Fehr & Fischbacher 2005) Punish opportunists (Fehr & Gachter 2002, Fowler & Christakis 2010) Keep groups relatively small (Stewart & Plotkin 2016) These contribute to social norm formation, which can stabilize cooperative behavior, and is commonly thought to be the basis of many religious and political traditions (Bohem 2011, Haidt 2012)

Elinor Ostrom & the Core Design Principles 1. Clearly Defined Boundaries 2. Proportional Equivalence between costs & benefits 3. Collective choice Arrangements 4. Monitoring 5. Graduated Sanctions 6. Conflict Resolution Systems 7. Minimal Recognition of Rights to Organize 8. Polycentric governance

Institutions can stabilize cooperation Ostrom’s work discovered these principles in the field (Ostrom 1990) Substantial evidence exists that these principles are mechanisms that stabilize cooperation in common pool resource management situations (Cox et al. 2010) These principles have been found to be generalizable across multiple contexts, and are likely evolutionary in origin (Wilson et al. 2013) Formally stated “Co-operative Principles” have evolved within the Co-operative Movement (Waring & Lange Forthcoming)

Spreading the Principles and Practice Social Learning, Imitation, and Cultural Group Selection have been the key factors in the spread of humans throughout all terrestrial biomes (Heinrich 2016). Using this knowledge allows us to spread the principles, and other findings of cooperation science, to improve the efficacy of groups.

To go forward We must assess where we are!

Buying Club Purchase Data ● Order data : who bought what ● Orders occur every week to every month ● 2012 - 2017 ● 27 Buying Clubs ● Clubs range from 10-20 members to almost 100 members

Buying Club Purchase Data What can it tell us? ● Item Details ● Item Cost ● Individual preferences ● Individual expenditures ● Splits - shared purchases ● Social network structure of splits ○ … all over time Item User Description Units per Case Cost per Unit Units Taken Amount paid Carrots by # 50 $1.50 10 $15.00 Bag of Carrots n/a $5.00 1 $5.00

Buying Club Survey Survey questions: - Broad: - Why do people join buying clubs? - What makes buying clubs fail or succeed? - Specific: - How cooperative are food buying clubs (compared to other groups)? - How does club structure relate to cooperation?

Survey Design & Implementation - Qualtrics online survey via BCS - Nov 2016 - March 2017 - 25 - 70 questions (members & coordinators) - Member perceptions & experience - Club history & organization (IDPs) - Demographics - Sample: 98 - 9 clubs

Survey Results Ordering - 54.3% purchases shared - 50% of coordinators believe more sharing would benefit the group Preferences - 33% joined to save money on food - 84% joined to support local producers Demographics - Mean HH income: 65,000, HH size: 3 - 63% bachelor’s or higher - > 95% female

Experimental Design One-shot dictator game (altruism) Public Goods Game (cooperation) - Presented at the start of the survey - Experimental economic games played with real $ (Square Cash)

Experimental Results Average donation among groups for each game - Mean dictator game: 58% of endowment - 107% increase from typical (Engel, 2011) - Mean public goods game: 66% of endowment - 74% increase from typical (Zelmer, 2003) - Elevated levels of cooperation when compared to other study populations (support for H1)

Summary Conclusions - Elevated levels of cooperation suggest that food buying clubs may promote cooperation, important for sustainable food systems - Clubs may have transformative effects by fostering prosociality Limitations - Small sample (98 obs) - Selection bias (online survey)

Upcoming Analyses ● Time series regressions ● Agent-based simulations ● Social network analysis

Time Series Analysis Can high split ratios be sustained? Is there split threshold below which clubs die?

Buying Club Simulations ● Cooperation maintained by reciprocity (Trivers, 1971). ● Generalized reciprocity between members on shared purchases (splits) ● Agents record the ratio of the help they’ve received versus given. ● Use ratio to determine whether to join help out.

Buying Club Simulations Cooperative momentum Generalized reciprocity Out of sample prediction? Fraction of Splits

Social Network Analysis What characterizes the splitting networks of successful or failing clubs? Are there structural early warning signs of failure?

Future Work ● Work with clubs to help achieve their goals ● Collaborations with co-ops organizations ○ National Cooperative Grocers ○ Cooperative Development Institute ● More groups, surveys, experiments, models ● More food!

Conclusions ● Cooperation is probably for cooperative organizations, and maybe for many other types of organization. ○ But, there is danger in relying on cooperation. ○ This has implications for local food in Maine. ● Cooperation and supporting institutions both evolve in groups. ○ And, we can measure it and nurture it.

Thank you! localfoodscience.org