Picking Winners When are technologyspecific policies optimal Discussion

Picking Winners: When are technology-specific policies optimal? Discussion by Brant Walker

Overview • Main idea: social planner takes product-level redundancy into account when making her decision • Under equal viability and heterogeneity between technologies • Equal Viability: Social Planner increases social welfare • No knowledge: Social Planner chooses a subset and can increase social welfare • Depends on heterogeneity between technologies • Knowledge: Social Planner chooses a smaller subset and increases social welfare • Social Planner can increase social welfare in winner-take-all

Strengths • Relevant in our current moment • Operation Warp Speed: government officials are “picking winners” for COVID-19 vaccines • Plans to support 8 vaccine candidates • Allows for heterogeneity across technologies and knowledge levels for social planners • Main reason why picking winners is not popular is because social planners do not have private information • The main result is consistent in a variety of scenarios

Questions • Do patent laws have an impact on product-level redundancy? • If a technology is too similar, may not be pursued • Potentially leads to less product-level redundancy • Is there a way to implement costs? • Assuming hiring scientists without labor costs (or costs in general) simplifies scientists’ decision • What market imperfections are assumed to not be present? • Imperfect competition, coordination failures between sectors, knowledge spillovers, information asymmetries, and environmental externalities • If no one has perfect information about technology viability, do we go back to the commonly held beliefs?

Suggestions • Instead of assuming equal viability among technologies, create an innovation possibilities frontier • • R(t) is the R&D expenditure which includes the “wage” for scientists • Possible Extension: May be interesting to understand how many resources are needed based on the amount of heterogeneity among technology viability and information

Schooling and Sustainability: What does Education tell us about Diversion Rates in Minnesota Counties Brant Walker, University of Iowa Jeff De. Simone, University of Iowa

Overview • Question: Is educational attainment related to observable sustainable behaviors, and if so, what levels? • Answer: Yes, bachelor’s degree level educational attainment is strongly linked with observable behavior • Fairly simple differences-in-differences two-way fixed effects • Identify causality as best as our data allows

Motivation MSW generation rates across the United States https: //www. epa. gov/facts-andfigures-about-materials-waste-and-recycling/national-overview-facts-and-figuresmaterials#Landfilling

Motivation • Education has positive effects on income, health outcomes, lower crime, job stability • Modern economics literature tends to agree that education is positively associated with sustainable behavior • Despite vast returns-to-schooling literature, there is a lack of causal evidence of this relationship • Our paper seeks to address this gap in knowledge

Specific Literature Review • Meyer 2015: Took advantage of European minimum schooling law change to find evidence of impact on sustainable behavior • Find increase in secondary-schooling requirements lead to an increase in 7 of the 8 pro-environmental behaviors considered • Chankrajang, Muttarak 2017: Instrumented for educational attainment through teachers per 1, 000 children in Thailand • Find increase in primary and secondary-schooling leads to an increase in some sustainable behaviors that are cost saving, but no increased concern about global warming or willingness to pay an environmental tax

Contribution • Our paper is complementary to the studies above because. . • Framework allows us to examine impacts across the distribution of educational attainment • Use U. S. data, where residents tend to have differing preferences to those studied previously • Study waste diversion, an all-encompassing waste management variable • Incorporates sustainable practices beyond recycling • Based on observed amounts of waste, rather than qualitative recycling behavior from self-reports

Data Strategy • Create new data set combining 5 -year averages of Minnesota SCORE and American Community Survey data from 2005 -2009 and 20102014 • 2 time periods • ACS data only reports schooling levels for geographic areas with populations under 20, 000 as 5 -year averages • Allows key variables such as education and income to change over a longer time frame, and mitigates measurement error in small counties • Main advantage: ability to examine relationship in a panel data framework with two-way fixed effects

Data •

Minnesota SCORE

Data • American Community Survey • Provides controls for other correlates of education • Per-capita income, total population, gender, race/ethnicity, age, employment, unemployment, and share of workers by occupation category and preferred transportation method • Main education measures • Particularly, percentage of residents of each county age 25+ with at least a bachelor’s degree • Other education measures are added in robustness checks

Empirical Strategy •

Main Model

Numerical Estimates • Back of the envelope • 1. 7% average increase in % of adults with bachelor’s degree • Educational effect is about 5. 25% increase in diversion rate

Alternative Models

Addressing Impact of Population

Population Robustness

Threats to Causality • Main limitation: No quasi-experimental mechanism • Put simply, it is possible that individuals who are more likely to attain a bachelor’s degree are also more likely to engage in sustainable behavior due to factors that influence both outcomes • Despite this, we show a significant impact of education on waste diversion practices persists through controls for labor market correlates such as… • Income • Employment, Unemployment • Occupational mix

Conclusion • Have shown 4 -year college graduation is a significant determinant of an area’s sustainable behavior • Could show through elected officials, businesses, and/or household habits • Cannot conclusively claim causality, but analysis identifies causality to the extent the data allows • First paper to utilize waste diversion data as a general measure of sustainable behavior • First paper to provide evidence in fixed effects panel data framework for a relationship between higher education and sustainable behavior

Thank you! I appreciate all comments Brant Walker University of Iowa brant-walker@uiowa. edu