Working paper

Kunxin Zhu, Sathya Gopalakrishnan, Daniela Miteva. Mangroves Help Reduce the Impact of Climate-induced Cyclones in India

Details

Mangroves shelter coastal economic activities during natural hazards like cyclones. However, despite consistent emphasis on such hazard mitigation benefits in mangrove preservation and restoration projects, there is insufficient information about the magnitude of such benefits, especially in India. Using night lights data, we quantify the cyclone mitigation impact of mangroves by analyzing 27 cyclone events that affected the eastern coast of India between 2012 and 2019. We estimate the overall short-term impact of cyclones on economic activity at the village-level and find that an hour of exposure to a category-4 equivalent cyclone decreased short-term local economic activity, proxied by night light intensity, by 12% for the average rural village along the east coast of India. Controlling for the endogenous distribution of mangroves, we find that on average 7% of cyclone damage was mitigated by mangroves. Second, the significant heterogeneity in the effect of cyclones across villages imply that natural hazard mitigation benefits from mangroves can substitute for local private investment in risk mitigation effort (e.g., quality of housing structures), as the marginal value of benefits from mangroves decreases in wealthier regions. Third, we find negative spillover effects in in adjacent villages, suggesting that migration into mangrove-protected regions may overestimate the cyclone mitigation benefits of mangroves.

---

Kunxin Zhu. Waves of Change: Understanding the Impact of Natural Hazards on Housing Market Dynamics

Details

As climate change intensifies, the frequency and intensity of natural hazards are expected to increase. Paradoxically, natural hazards can trigger residential redevelopment, increasing the footprint of built capital and placing a higher value at risk for future hazards. By examining the spillover effects of private investment in renovation, this work enhances our understanding of the impact of natural hazards on the patterns and dynamics of durable housing capital investment along coastal areas. Using parcel-level information on property tax assessments and decomposed property values, I find that Hurricane Irma led to a significant increase in low-cost maintenance, with an 80% rise in the annual probability of renovation on main buildings or patios for affected properties. However, I do not find evidence that Hurricane Irma increased the likelihood of post-disaster reconstruction. Using disaster-induced renovations as an instrument, I find that maintenance in the neighborhood increases the likelihood of non-affected properties undergoing renovation. Employing a durable property investment model that allows for both maintenance and replacement, I conceptually demonstrate the reasons for heterogeneous impacts of natural hazards on subsequent renovation and reconstruction decisions. Results from calibrated models indicate that communities facing repetitive flood risks tend to overinvest in low-cost maintenance, which delays reconstruction behaviors while improving overall housing quality in natural hazard-prone areas.

---

Kunxin Zhu,Sathya Gopalakrishnan, Martin Smith. Disaster-induced Migration Across U.S. States: The Role of Income Heterogeneity

Details

While climate change is increasing the frequency and intensity of natural disasters in coastal areas, more people are choosing to live in coastal areas. These trends are unfolding against a backdrop of rising income and wealth inequality in the United States. Disasters have direct effects on inequality when low-income households face higher risk exposure. Effects can also be indirect when disasters trigger migration that affects the resulting community composition. Although the distributional impacts of climate change are widely acknowledged, few empirical studies examine disaster-induced migration as a potential mechanism driving inequality. We analyze the effects of disasters on migration and show theoretically the potential for an inverted-U shape: low-income households are unable to move to other locations because of financial constraints, high-income households can adapt in place with mitigation investments such as elevating housing structures, and middle-income households are, therefore, most likely to migrate. We empirically quantify the extent to which migration response is heterogenous across income strata. Using a household-level dataset covering coastal counties in the Atlantic and Gulf coasts of the U.S., we find an inverted-U relationship between household economic resources (income and wealth) and the effect of natural disasters on out-migration decisions. The results are consistent with our theory that households with different economic resources use different adaptation strategies in response to natural disasters, suggesting long-run demographic change in natural disaster prone regions. Specifically, as middle-income households continue to migrate out because of an increase in future natural disaster events, we expect to see a higher concentration of low-income and very high-income households in the Atlantic and Gulf coast regions and further exacerbated income inequality. Our findings add to the literature on the distributional effect of climate-induced migration and indicate the equitable “migration as adaptation” outcome cannot be achieved in the current political-economic setting.

---

Kunxin Zhu,Sathya Gopalakrishnan. The Impact of Natural Hazards on Local Communities: Understanding Changes in Homeownership Rate after Floods in Iowa

Details

Significant changes in housing markets result from natural disasters increasing in the frequency and intensity. While research has acknowledged that flooding can have heterogeneous effects on homeowners and renters, few studies have analyzed the effect of flooding on homeownership rates. Using the panel fixed effect model, we find that flooding events can decrease average homeownership rates by 1.2% in Iowa. Using a major flood in 2008 as a natural experiment, we decompose the effect of migration responses to the flood and find that the negative effect on homeownership can be persistent for over 6 years in inundated areas. We also find that the decrease in homeownership rate is driven largely by outmigration of existing owners. Our findings add to the literature on the effect of natural disasters on housing markets and inform policymakers about major channel for this transition.

---

Work in progress

Kunxin Zhu and Yunan Zhu. Building back better? The effect of post disaster assistance on housing development.

Daniela Miteva,Sathya Gopalakrishnan, Sammuel Cheng, Kunxin Zhu, Andrew Miller.The marginal value of forests in rural India.