Do remittances make poor households more resistant to ‘natural disasters’? Evidence from the 2015 earthquake in Nepal
Introduction
Studies have shown that so-called ‘natural disasters’ arise when socio-economic factors such as poverty and political dysfunctions collide with natural hazards (e.g., Refs. [[1], [2], [3], [4]]). This view led to the hypothesis that one efficient measure to mitigate ‘natural disasters’ is to reduce poverty. Empirical support for this hypothesis is in general solid among macroeconomics studies with cross-country panel data. For example, Anbarci et al. [5],Kahn [6], and Toya and Skidmore [7] estimated negative associations between per capita GDP and the number of deaths from natural hazards. The use of macro data, however, masks the channels through which a rising income mitigates ‘natural disasters’.
Another strand of literature in hazard-reduction research has tried to determine the factors that motivate a household (HH) to prepare against natural hazards (e.g., Refs. [[8], [9], [10]]). Many such studies distinguish two types of ‘HH preparedness’ following Lindell and Perry [11]. One type is hazard mitigation: passive self-protection against nature's shocks. Typical examples are constructing houses with stronger materials (e.g., reinforced concrete against earthquakes) in safer locations (e.g., higher ground against hurricanes). The other type is emergency preparedness: active responses for survival after hazard impacts are experienced, a typical example of which is stockpiling emergency supplies at home. Using HH data, many studies in this field have found positive associations between HH income and HH preparedness for different types of natural hazards: e.g., Chaney et al. [12] for tornadoes, Donner and Lavariega-Montforti [13] for floods, and Lindell and Perry [11] and Tekeli-Yesil et al. [14] for earthquakes.1 Their findings suggest a main channel through which income mitigates negative impacts from natural hazards: a higher income enhances HH preparedness, and the preparedness effectively absorbs the impacts of nature's shocks.
These two associations become, however, less clear in the analyses focusing on developing countries. Regarding the negative association between per capita GDP and the damages incurred from natural hazards, Toya and Skidmore [7] pointed out that the estimates are much smaller in developing countries compared to those in OECD countries. Kellenberg and Mobarak [15] argue that nature's shock mitigation is a luxury good among the poor in developing countries, and indeed estimated that fatality rates from flooding, landslides, and windstorms increase with per capita GDP before they decrease: inverted U associations. As for the positive association between a higher HH income and HH preparedness for natural hazards, there are not yet many studies conducted in developing countries.2 A few exceptional studies such as Ashenefe et al. [16] found a positive association. Hoffmann and Muttarak [17] and Samir [18], however, suggest that income is not an important factor to improve HH preparedness for natural hazards.
This paper contributes to the literature on these two associations in low-income countries. By utilizing detailed HH survey data from rural areas of Nepal, we investigated whether a higher income mitigated the damages caused by the 2015 devastating earthquake to houses. We pay special attention to the effects of remittances, because in many developing countries, remittances from temporary migrating family members constitute a large part of rising HH income (e.g., Refs. [19,20]). Our analysis consists of two steps. First, we examine the association between HH income and houses with concrete walls. This part essentially provides a case to support the literature on the association between HH income and hazard mitigation behaviors in low-income countries. Second, we estimate whether HH income and houses with concrete walls are negatively associated with structural damage from the 2015 seismic shocks. Unless HH preparedness effectively absorbs the nature's shocks, it cannot be considered a major mitigation channel through which a higher income results in smaller hazard damages.
On the relationship between building structures and seismic damages, there are a vast number of earthquake engineering studies (specific to the 2015 earthquake in Nepal, refer to, e.g., Refs. [[21], [22], [23], [24], [25]]. Few studies, however, have explicitly examined the link among HH income, house structure, and actual damage from natural hazards. The exceptional studies such as Lim et al. [26] and Milch et al. [27] used aggregated data at local-administration units. With the detailed HH data including, for instance, income variables differentiating sources, we attempt to make a clear case for the three-body link among HH income, construction materials of house structure, and damage from hazards.
It has been well-documented that Nepal has both a large migrant worker population abroad as well as a high rate of remittances [28]. Mohapatra et al. [29] examined the associations between remittances and hazard preparedness, and found that in Ghana and Burkina Faso, those HHs receiving remittances were more likely to have houses built of concrete rather than mud and bricks. In Nepal, with the original data of 798 HHs in urban areas collected in 2014, Manandhar [30] showed that remittances drove construction of new houses with concrete structures. Additionally, he pointed out that in spite of the requirement in the national building code of Nepal, remittance-dependent HHs in his data set often constructed new houses without hiring architect engineers. The findings of Manandhar [30] raised the concern of whether the concrete houses built using remittances were resistant to the 2015 seismic shocks. In fact, Liu et al. [22] described that most of self-built reinforced concrete frame structure collapsed partially or completely. Few existing studies have examined whether HH hazard mitigation choices, in particular those in developing countries based on remittances, effectively reduce damage from natural hazards.3 We intend to provide such an analysis.
This paper proceeds as follows. The next section summarizes the background and the data utilized in this study. One advantage of our study is that the data include a house damage measurement objectively assessed by architect engineers. One limitation lies in the relatively small sample size: 110 HHs. Based on our analytical framework put in Appendix B, Section 3 details our empirical strategy. Section 4 reports the results and Section 5 provides a discussion on them. We found a positive association between remittances and the use of concrete for houses and between remittances and house damage due to the seismic shocks. However, we were unable to find any statistical evidence to support the association between house structure (i.e. concrete walls) and damage to houses.
Section snippets
The 2015 Nepal earthquake (NEQ)
In 2015, a massive earthquake on April 25 with an estimated 7.8 moment magnitude and aftershocks struck Western and Central Nepal [47]. The Government of Nepal (GoN) declared that 31 out of the total 75 districts of the country were impacted by the earthquakes (Fig. 1). Among them, 14 districts were categorized as ‘severely or crisis hit districts’. The number of collapsed private houses amounted to 351,808, and overall 792,987 houses were categorized as seriously damaged and in need of
Variables
Our analytical framework is put in Appendix B. Income is the first critical variable that our framework suggests for empirical analyses on the use of concrete for walls (Ho in Appendix B) and maintenance and minor repairs of the house for preparing against earthquakes (Hu). Our data set can differentiate income into remittances (R) and cash income from local sources (Y). The second critical variable is the preference for insuring against seismic risks (θ). We cannot directly observe θ, but
Summary statistics
Table 1 shows the number of HHs in the three Village Development Committees (VDCs) from the 2011 Population Census of Nepal, and our samples in these VDCs. We refer to the VDC-level numbers because village-level information is not available in the 2011 Census. Due to temporary migration of working-age males, the male population ratio is smaller than 45%: a typical number in rural areas of Nepal. The average HH size and the ratio of male population indicate that our sample HHs are similar to the
Discussion
In this study, we explored the associations between rising household (HH) income and damage from natural hazards in low-income countries. By utilizing the detailed HH survey data from the rural areas of Nepal, we sought to elucidate the three-body link among HH income mainly from remittances, house structures, and the damages to house structures during the 2015 devastating earthquake. A major advantage of this study is that our data contain a house damage index that was objectively evaluated
Conclusion
Remittances make up a sizable income source in many developing countries. The results of this paper suggest that rising incomes through remittances do not necessarily reduce so-called ‘natural disasters’. More specifically, simply using modern and supposedly seismic resistant construction materials with rising incomes does not necessarily keep HHs safe from nature's shocks. To mitigate ‘natural disasters’, rising HH incomes must be accompanied by improved social infrastructures including
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We thank the National Reconstruction Authority of the Government of Nepal for giving us an access to the damage-evaluation data of houses collected for the ‘Household Registration for Housing Reconstruction Program’. We also thank the International Food Policy Research Institute for providing us with the data of Kumar and Hotchkiss [31], two anonymous referees of the journal, Dr. Naohito Abe and Dr. Masahiro Shoji for providing us useful comments. This research was supported by JSPS KAKENHI
References (51)
- et al.
Earthquake fatalities: the interaction of nature and political economy
J. Publ. Econ.
(2005) - et al.
Economic development and the impacts of natural disasters
Econ. Lett.
(2007) - et al.
Household preparedness for natural disasters: impact of disaster experience and implications for future disaster risks in Japan
Int. J. Disaster Risk Reduc.
(2017) - et al.
Does rising income increase or decrease damage risk from natural disasters?
J. Urban Econ.
(2008) - et al.
Learn from the past, prepare for the future: impacts of education and experience on disaster preparedness in the Philippines and Thailand
World Dev.
(2017) - et al.
Reflection on earthquake damage of buildings in 2015 Nepal earthquake and seismic measures for post-earthquake reconstruction
Structure
(2021) - et al.
Field investigation on the performance of building structures during the April 25, 2015, Gorkha earthquake in Nepal
Eng. Struct.
(2016) - et al.
Seismic performance of buildings in Nepal after the Gorkha earthquake
- et al.
Double danger in the double wide: dimensions of poverty, housing quality and tornado impacts
Reg. Sci. Urban Econ.
(2017) Remittance and earthquake preparedness
Int. J. Disaster Risk Reduc.
(2016)