Prevent or repair? Experimental evidence from providing incentives for climate resilient housing in Vietnam

ABSTRACT

We present results from a randomized field experiment in Da Nang in Central Vietnam. We assess the take-up and impacts of a microcredit program that aims to increase the adoption of climate resilient housing among low-income urban households. Households were randomly assigned offers of either a loan and technical assistance package or a cash transfer with a smaller loan and technical assistance. We find large and significant impacts on the resilience of the new or retrofitted houses due to being offered the more generous incentive package. Households that were offered the cash transfer and a smaller loan, are three times as likely to accept. The difference in uptake, and therefore impact, of the two packages shows that there is a need for a subsidy in addition to the technical assistance in order to reach near-poor households. Previous public spending on typhoon relief indicates that such a subsidy could be covered by re-allocating funds from typhoon repairs to damage prevention through resilient housing.

KEYWORDS:

• Resilient housing
• randomized control trial
• climate change adaptation
• urban
• Vietnam

Acknowledgements

We would like to thank the Women's Union of Da Nang for their cooperation and hospitality, as well as The Institute for Social and Environmental Transition (ISET) – Vietnam. We are grateful to the City of Da Nang and all households participating in the project. An early version of this paper received useful comments from participants at the 16th Nordic Conference in Development Economics and from Henrik Lindhjem. We also thank seminar participants at CICERO Center for International Climate Research for useful feedback. All remaining errors are our own.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 A typhoon is a rotating system of clouds and thunderstorms with wind speeds exceeding 74 miles per hour, originating in the Northwest Pacific region (http://www.nhc.noaa.gov/climo/).

2 Associated Press, https://usatoday30.usatoday.com/weather/storms/\Clinicaltrialid{\Clinicaltrialid-10-03}}-asia-typhoonx.htm

3 http://blog.i-s-e-t.org/we-are-honored-to-accept-the-unfcccs-lighthouse-activities-award/

4 Hudson et al. (2019) show that flood impacts on subjective well-being of residents in Central Vietnam may be important and should be accounted for in risk management strategies.

5 Wards and communes are the same administrative unit, but the name differs between rural areas (communes) and urban areas (wards).

6 We have included the 8 wards in which the WU implemented their Rockefeller Foundation supported project from 2011 to 2016. In our analysis of take-up and short run impacts of the current program, we include a dummy variable for the wards that were included in the Rockefeller project to control for this, and to investigate whether this affects take-up or impacts.

7 The Da Nang City poverty lines were updated in 2016 (Da Nang City People's Council, 2015): Urban poor households: Income per capita per month (ICM) $<1300000$ VND, Urban near-poor households: 1300000 VND ≤ ICM $\le 1690000$ VND, Rural poor households: ICM $<1100000$ VND, Rural near-poor households: 1100000 VND ≤ ICM $\le 1430000$ VND. This corresponds to a near-poor poverty line of about 75 USD per capita per month in urban wards, and 63 USD per capita per month in rural areas of Da Nang. By comparison, the national poverty line for urban areas was set at 760,000 VND per capita per month for urban areas, and 615,000 VND per capita per month for rural areas, according to the General Statistical Office of Vietnam (General Statistics Office of Viet Nam, 2016).

8 Around 30% of households in urban areas of Vietnam were female-headed in 2016 (GSO, 2016b).

9 Government planning zones are areas that may be included in city plans for infrastructure projects or other projects.

10 As a robustness check, we also compute p-values using the wild bootstrap procedure described in Cameron et al. (2008), since the relatively small number of clusters may be a problem for standard cluster-robust standard errors. The significance levels are largely robust, with exceptions noted in the text.

11 Wild bootstrapped standard errors give slightly higher p-values, from 0.005 to 0.08 for the coefficient on package 1*income, and from 0.008 to 0.01 for the coefficient on package2*income

12 Note that all households have expressed an interest in retrofitting or reconstructing their house to ensure storm resilience, regardless of previous program exposure. This suggests that the finding is not because these households have already invested in storm resilient housing.

13 The results are robust to using wild bootstrapped p-values.

14 Since roof bracings are only used with corrugated steel sheet roofs or clay tile roofs it is not possible to have all six resilience elements present.

15 The results are robust to using wild bootstrapped p-values.

16 Reaching poorer households than the ones targeted in this program likely requires a larger grant element or a grant only for some households.

17 https://www.worldbank.org/en/topic/disasterriskmanagement/brief/global-program-for-resilient-housing

18 For instance, the Poverty Reduction Plan of Da Nang City for 2015–2020 (Da Nang City People's Committee, 2015) includes a budget plan for providing grants for more than 2500 retrofitted or reconstructed houses for poor households in the city.

Additional information

Funding

This research was supported by the Nordic Development Fund under the Nordic Climate Facility.

Notes on contributors

Sofie Waage Skjeflo

Sofie Waage Skjeflo is a senior researcher at CICERO Center for International Climate Research Oslo. She holds a PhD in economics from the Norwegian University of Life Sciences from 2015. Her research interests include empirical environmental and development economics, climate adaptation and climate mitigation policies.

Nina Bruvik Westberg

Nina Bruvik Westberg holds a PhD in economics from the Norwegian University of Life Sciences (NMBU) and a master’s degree in development and natural resource economics. She is currently placed at Menon Economics. Nina works on climate-, environment- and poverty-related issues, including governments’, companies’ and households’ strategies for mitigating and adapting to climate change.

Haakon Vennemo

Haakon Vennemo is a professor of economics and partner at the economics consultancy Vista Analyse. He holds a PhD from the University of Oslo. His research interests are development economics, environmental economics and public economics. He has contributed to the literature on co-benefits of climate mitigation action and the marginal cost of public funds, and a range of other topics.

Tuan Huu Tran

Tran Huu Tuan completed his PhD at the Norwegian University of Life Sciences, Norway. His specialities are Environmental economics and Sustainable tourism. Tuan has about 20 papers published in SCI/Scopus journals. He is currently the Rector of the School of Hospitality and Tourism - Hue University, Vietnam.

Phong Van Giai Tran

Dr Phong Van Giai Tran has been working as a practitioner in the field of climate change resilience and disaster risk reduction in developing countries in Asia and the Pacific, particularly in Vietnam, for more than 20 years. He has intensive knowledge of climate change adaptation and disaster risk reduction theories and practices. He is currently working for the Swiss State Secretariat for Economic Affairs (SECO) in Vietnam.

Tran Tuan Anh

Tran Tuan Anh obtained his master’s degree in 2008 at Chiang Mai University (Thailand) and his doctoral degree in 2015 at RMIT University (Australia). His expertise focuses on housing vulnerability, post-disaster housing reconstruction, and building community resilience with a focus on the Vietnam context. Tuan Anh is currently a lecturer and researcher at Hue University of Sciences (Vietnam).