This website uses cookies.

By clicking the "Accept" button or continuing to browse our site, you agree to first-party and session-only cookies being stored on your device to enhance site navigation and analyze site performance and traffic. For more information on our use of cookies, please see our Privacy Policy.

  1. Home
  2. ASSA Annual Meeting
  3. 2022
  4. Preliminary Program
  5. Behavioral Response to Environmental Interventions
« Back to Results

Behavioral Response to Environmental Interventions

Paper Session

Friday, Jan. 7, 2022 12:15 PM - 2:15 PM (EST)

Hosted By: Association of Environmental and Resource Economists
  • Chair: Kenneth Gillingham, Yale University

Pollution Monitoring, Strategic Behavior, and Dynamic Representativeness

Lin Yang
,
Hong Kong University of Science and Technology
Yatang Lin
,
Hong Kong University of Science and Technology
Jin Wang
,
Hong Kong University of Science and Technology
Fangyuan Peng
,
Hong Kong University of Science and Technology

Abstract

Air quality evaluation in major countries around the world is mainly based on stationary, in situ
monitors that aim to provide a representative measure of local air quality. To comply with air
quality standards, local governments may take targeted measures to achieve better monitor
readings. The strategic response could lead to changes in the ground monitoring system's spatial
representativeness in the long run. Using high-resolution satellite-based air pollution measures,
we examine local governments' strategic behavior and its implications on dynamic
representativeness based on the staggered roll-out of the monitoring system in China. The
analysis shows that local governments target pollution reductions in areas closer to monitors
after monitor installations, leaving pollution elsewhere unchanged or even increased. We also
find heterogeneities in local officials' strategic responses when they have different political
incentives, e.g., larger strategic reductions in cities with younger mayors. One potential channel
is through changes in local industrial activities. Using satellite-based measures of thermal
anomalies, we show that the intensity of industrial production activities in monitored areas
significantly decreases after monitoring. This suggests that local governments may have ordered
polluters adjacent to monitors to decrease their production to lower pollution readings. To shed
light on the regulation effectiveness, we show that most of the monitors are ex-ante good
representations of the city's average air quality. However, the local officials' strategic responses
have changed the spatial representativeness dynamically. Given that monitor locations are
unlikely to change once sited, the ground monitoring system will no longer be representative in
the long run. Our results suggest an improved policy design for air quality evaluations, which
needs a combination of ground monitoring data and auxiliary pollution information from remote
sensing data and public supervision.

The Effect of Rebate and Loan Incentives on Residential Heat Pump Adoption: Evidence from North Carolina

Xingchi Shen
,
University of Maryland-College Park
Yueming (Lucy) Qiu
,
University of Maryland-College Park
Pengfei Liu
,
University of Rhode Island
Anand Patwardhan
,
University of Maryland-College Park

Abstract

Electrification can promote deep decarbonization to tackle climate change with a cleaner power grid. Electric heat pumps provide a feasible and energy-efficient way to replace fossil-fuel furnaces for space heating. Rebate and loan programs are the two most-widely used incentives for residential heat pump installations in the US. Leveraging the geographical variation of incentives for heat pumps provided by different utilities in North Carolina, we estimate the effect of a rebate program on heat-pump adoption and compares the effect of the rebate with those of two loan programs via three different quasi-natural experiments. We examine a narrow buffer zone along the borderlines between the Duke Energy Corporation and three other electric utilities with different incentives (or without incentives). The borderline must be within the same ZIP code areas. The Duke Energy utility provides a rebate program and is adjacent to three other utilities (Rutherford EMC, Union Power EMC, and Haywood EMC). Rutherford EMC provides no incentives, Union Power EMC provides a loan program with 9% interest rate, and Haywood EMC provides a loan program with 3.9% interest rate for encouraging heat pump installations, which serve as three control groups, respectively. First, using difference-in-differences, our results show that the rebate program ($300-$450 per system) increases the adoption density by 0.049 in a year (or a 26% increase). Second, by comparing differential growth trends of heat pump adoption within a narrow buffer zone, we find that the rebate program is more effective in promoting heat pump adoption for average consumers than two loan programs (annual loan interest rate: 9%, 3.9%). Third, we find the rebate program is less effective for low-income households than high-income households. Last, we find the rebate program is also more cost-effective than the two loan programs if same proportions of residents apply for the rebate and the loan.

Using Targeting to Optimize Program Design: Evidence from an Energy Conservation Experiment

Todd Gerarden
,
Cornell University
Muxi Yang
,
Cornell University

Abstract

We investigate the potential for targeted treatment rules to improve the performance of a large-scale behavioral intervention to encourage households to conserve energy. We first verify that receiving personalized feedback on energy use causes households to reduce their electricity consumption on average, and that there is heterogeneity in treatment effects. We then exploit this heterogeneity by using a policy-learning approach to derive treatment rules based on observable household characteristics that maximize the expected benefits of the intervention. Targeting treatment by using transparent and easily implemented rules could yield significant gains; the energy savings from optimal treatment assignments are predicted to be double those achieved by the intervention as implemented. Predicted cost savings from targeting are even larger. Our results underscore the potential for targeted treatment assignment to generate large welfare gains in many domains.

Peer Effects in Residential Green Infrastructure Adoption

Daniel Brent
,
Pennsylvania State University
Joseph Cook
,
Washington State University
Gabriel Lara
,
Pennsylvania State University
Allison Lassiter
,
University of Pennsylvania
Douglas Wrenn
,
Pennsylvania State University

Abstract

Stormwater runoff can cause untreated stormwater and wastewater to flow directly into water
bodies during heavy rain events. These combined sewer overflows (CSOs) violate the Clean
Water Act, and mitigation costs billions of dollars. Many cities use voluntary green stormwater
infrastructure (GSI) on private land to meet obligated stormwater reductions, but adoption
decisions for private GSI are understudied compared to participation in other voluntary
environmental programs. We investigate peer effects in voluntary GSI adoption with a unique
identification strategy exploiting spatial and temporal eligibility in Seattle’s RainWise program.
Our identification strategy exploits households’ relative position in eligible sewersheds as
plausibly random variation in the number of potential peers. Households in the middle of an
eligible sewershed have more eligible peers than households at the border. Eligible sewersheds
are selected for hydrological priorities and sewershed boundaries are uncorrelated with
socioeconomic characteristics of the households or traditional neighborhood boundaries.
This exogenous variation in the number of eligible peers strongly impacts the number of peers
that ultimately adopt RainWise. Our IV specification estimates that an additional peer adoption
causes a 0.3% increase in the probability of signing up for RainWise, relative to a 0.4% baseline
annual adoption probability. The IV estimates are three times higher than naïve OLS estimates
highlighting the importance of identification in estimating peer effects. We argue that our
instrument is more likely to be exogenous than other instruments used to identify environmental
peer effects such as housing market turnover. Initial estimates using housing market turnover,
which weakly increases adoption, are insignificant and of the reverse sign. We also estimate
spillover effects of other environmental programs such as water conservation programs and find
mixed results. This research contributes to the literature on voluntary adoption, peer effects, and
green stormwater infrastructure for water quality.

Discussant(s)
Corbett Grainger
,
University of Wisconsin-Madison
Christine L. Crago
,
University of Massachusetts-Amherst
Christopher R. Knittel
,
Massachusetts Institute of Technology
Kenneth Gillingham
,
Yale University
JEL Classifications
  • Q4 - Energy
  • C4 - Econometric and Statistical Methods: Special Topics