Adaptation Through Tile Drainage: A Structural Ricardian Analysis (w/ David A. Keiser)
This paper provides the first estimates of the effects of climate change on agriculture while explicitly modeling tile drainage. I show in a simple conceptual model that the value of precipitation differs between drained and non-drained land, implying that pooling these lands could bias estimates of the effects of climate change on land values. I test this hypothesis by estimating a Structural Ricardian model for U.S. counties east of the 100th meridian. Consistent with the theoretical model, the estimates show that the value of precipitation is higher on non-drained lands. When tile drained land is accounted for, estimated damages from climate change are consistently higher than pooled estimates across model specifications.
Land Use Change and Lake Water Quality: A Dynamic Panel Data Approach
Understanding how land use change affects water quality is critical to evaluating the full welfare effects of policies such as the biofuel mandate and the Conservation Reserve Program. Many water quality studies rely on cross sectional data or time series analysis of a single lake. I combine water quality measurements from over 120 Iowa lakes over 10 years with satellite data on land use change to provide new estimates of this response. Panel data allows for control of fixed effects, which mitigates potential omitted variable bias. A dynamic model captures the stock effect of pollution in lakes over time. I use the coefficients to estimate the effect of the biofuel mandate on lake water quality as well as the welfare effects.
This paper illustrates how to improve the immersiveness of an environmental valuation study using virtual reality headsets and real video footage. Recent research has used ``virtual environments" to study this issue, however technological advances in virtual reality headsets allow for a far greater degree of immersion. In this study, subjects were randomly shown either a virtual reality video or static pictures of a polluted lake, before and after cleanup. They were then asked to indicate whether they would be willing to pay a random amount to improve lake water quality to the level shown. A discrete choice model is used to estimate and compare the willingness to pay for both groups. In this case study there was no detectable effect on willingness to pay estimates. However, the technology may be beneficial for other valuation scenarios, particularly when the environmental change is complex or difficult for participants to evaluate.
The Effect of Laptop Use on a Neighbor’s Learning
Although laptops are ubiquitous in college classrooms, they can produce a negative externality by distracting neighbors and impeding the learning process. Previous research that has studied computers in the classroom have relied on quasi-experimental designs or lab experiments . This paper uses a different approach by analyzing the effect in real classrooms over an entire semester. The key innovation is that students are randomly assigned a seat during each lecture, effectively randomizing laptop locations, and therefore laptop exposure, throughout the classroom. At the end of each lecture students were given a quiz on that day's material, which is used to test the hypothesis that increased exposure to neighboring laptops lowers the average quiz score. A simple regression shows a negative correlation, however once fixed effects are included, the effect vanishes. This provides evidence that laptop use is not detrimental to neighboring students as indicated by previous studies.