« Back to Results

Environment and Health

Paper Session

Saturday, Jan. 5, 2019 2:30 PM - 4:30 PM

Atlanta Marriott Marquis, M202
Hosted By: Association of Environmental and Resource Economists
  • Chair: Jonathan Ketcham, Arizona State University

Does When You Die Depend on Where You Live? Evidence from Hurricane Katrina

Tatyana Deryugina
University of Illinois-Urbana-Champaign
David Molitor
University of Illinois-Urbana-Champaign


Hurricane Katrina devastated the Gulf Coast in 2005, displacing over 1 million
people. We follow Medicare cohorts over time and space to estimate the hurricane's long-run mortality
effects on elderly and disabled victims who were initially living in New Orleans. We estimate that the
hurricane reduced long-run mortality: inclusive of the initial shock, victims are 1.75 percentage points
more likely to be alive eight years after the storm. Two patterns indicate that migration to lowermortality
regions drives this mortality reduction. First, victims in flooded neighborhoods migrated at
much higher rates and experienced greater mortality reductions. Second, although migrants who moved
to regions with lower mortality look similar at baseline to those who moved to higher-mortality regions,
migrants' subsequent mortality is 0.98-1.12 percentage points lower for each percentage-point
reduction in local mortality. By contrast, movers' subsequent mortality is unrelated to local Medicare
spending. On average, Hurricane Katrina victims relocated to lower-mortality areas, which explains 56-
79 percent of the overall mortality reductions we find.

Opening the Black Box of Information Interventions: Evidence from Environmental Health Practices in India

Emily Pakhtigian
Duke University
Subhrendu Pattanayak
Duke University


Environmental protection is difficult partly because of the peculiarities of
environmental contamination – households are unaware of the extent contamination or of
averting behaviors. Thus, information interventions are a common policy response. Given the
low rates of sustained adoption of many environmental health technologies, we posit that it is
key to open the black box and describe the information and belief forming mechanisms. In this
paper, we adapt a model of impure public goods and draw on a decade-long follow up to a
sanitation information experiment in Odisha, India. This setting is important because India is a
hot spot for sanitation as rates of open defecation remain around 40 percent in 2015, despite
concentrated governmental efforts to promote latrines. Using a panel of approximately 1100
households collected over an entire decade (measurements in 2005, 2006, 2010 and 2016), we
empirically test the channels through which the randomized information experiment operated
in the short and long terms. Specifically, we consider how knowledge of sanitation-disease
linkages, perceptions of village cleanliness, and beliefs about the acceptability of sanitation
practices change as a result of exogenous information receipt and whether these updates are
sustained over time. We find that each indicator improves immediately following the
intervention; however, the short-term differences between treatment and control villages do
not persist at the ten-year mark. Furthermore, we consider how this updating process relates to
household decisions to adopt (or abandon) latrines, taking into account social, health, and
preference-driven benefits and costs of adopting improved sanitation technology. We find that
trends in knowledge and beliefs largely align with trends in latrine adoption and abandonment
found by previous studies, suggesting the policy importance of these mechanisms.

Environment-Enhanced Momentum and the Demand for Environmental Quality

Haoming Liu
National University of Singapore
Jingfeng Lu
National University of Singapore
Alberto Salvo
National University of Singapore


We infer the demand for environmental amenities from strategic behavior in dynamic contests. We develop a best-of-three dynamic contest model of complete information to investigate how environmental factors shift worker payoffs and effort in equilibrium. The model accommodates player asymmetry and randomness, and can allow for a “psychological” effect of winning and other forms of serial correlation. The model predicts that high temperatures and particle (PM2.5) pollution enhance the strategic momentum effect, and may give the weaker player a better chance overall. We implement the model on a sample of professional tennis matches in Australia and China and obtain strong evidence of environment-enhanced momentum. A world-class athlete’s willingness-to-pay to avoid playing in hot weather or polluted air is $7,000. We discuss implications for labor supply given a warming climate, as well as applicability to a wide range of economic settings.

Environmental Protection for Bureaucratic Promotion: Water Quality Performance Reviews of Provincial Governors in China

Liguo Lin
Shanghai University of Finance and Economics
Wei Sun
Shanghai University of Finance and Economics
Jinhua Zhao
Michigan State University


Many developing countries inadequately enforce their environmental regulations. Economists have
studied a variety of reasons underlying the imperfect enforcement, including asymmetric information,
capacity constraints, corruption, etc. In this paper, we study one reason that has not attracted much
attention: regulators’ incentives for bureaucratic promotion.
We empirically demonstrate that, by simply adding water quality targets in the performance
review of provincial governors, the Chinese government has reduced ambient water pollution and the
associated digestive cancer death rate, although these come at the cost of slower economic growth.
Using a unique data set that matches different water quality measurements with death by cause data,
and taking advantage of the gradual expansion of the water quality performance review (WQPR) over
time and space, we estimate the effects of the policy change on ambient water quality, digestive cancer
mortality, and the GDP growth rate.
We find that WQPR improved water quality attributes that are included in the performance
review but not attributes excluded from the review. The improvements manifest not only in the
monitoring data reported by local environmental agencies but also in the data collected directly by the
central government, ameliorating concerns for data manipulation. WQPR reduced the mortality rate of
digestive cancers but not of cancers unrelated with water quality such as lung cancer. The
improvements are higher at provincial boundaries, which are specifically targeted by WQPR, and are
higher when the governors have higher promotion potentials. As typical of command-and-control
policies, WQPR comes at significant economic costs, reducing short-term GDP growth rates by 1.14%.
Our findings demonstrate the importance of incentivizing regulators for enforcement. In
developing countries with bureaucratic systems, a strong incentive lies in the performance review and
promotion of government officials. Regulatory design should thus target not only polluters but also
Jonathan Ketcham
Arizona State University
Fiona Burlig
University of Chicago
John Wooders
New York University Abu Dhabi
Guojun He
Hong Kong University of Science and Technology
JEL Classifications
  • Q5 - Environmental Economics
  • I1 - Health