Research
Do Disasters Always Lead to an Increase in Intimate Partner Violence? Evidence from the 2018 Earthquake in Papua New Guinea
with Alyssa Leng
We examine how an earthquake in Papua New Guinea changed attitudes about and the prevalence of intimate partner violence (IPV). Although there are several reasons why disasters can aggravate IPV, the acceptability of IPV amongst men in disaster-affected regions significantly declined. There was a smaller and noisier decline in reported incidents of IPV, driven by declines amongst women least likely to underreport IPV. The results highlight that the responsibilities of household members and social norms can change in sufficiently turbulent disasters, which can lead to IPV improvements, and that measurement issues need to be better addressed to improve understanding of IPV.
with Anna D'Souza, Eliana Favari, and Siddharth Krishnaswamy
World Bank Policy Research Working Paper WPS10176
Revision Requested (Economic Development and Cultural Change)
We investigate the causes and consequences of forced displacement using a survey that reached households before and after becoming displaced during an active conflict. Previous work has only been based on data from households after becoming displaced and has thus been unable to more formally investigate the causes of forced displacement, has been unable to directly estimate the impacts of forced displacement on households, and has been unable to estimate the persistence of those impacts. Using this novel data, we find that there were no sharp drops in food access prior to displacement, that displacement resulted in an immediate decline in food access, and that food access quickly rebounded to pre-displacement levels. We further investigate potential explanations for the quick rebound in food access, including the provision of emergency food assistance. These estimates illustrate that displacement is not always being forced by sudden changes in humanitarian conditions, and that critical dimensions of well-being can rebound quickly.
How Much of Economic Growth Trickles Down to the Population in Resource-Rich Countries? Evidence from Papua New Guinea
with Paripoorna Baxi and Darian Naidoo
There has been substantial growth in the resource sector in PNG during the last resource boom, increased revenue collection by the government associated with that growth, and significant increases in international assistance, all which might have translated into improved well-being outcomes across the country. To better understand whether these changes improved household-level outcomes, we update estimates of key well-being outcomes in the country. Specifically, we impute monetary poverty status using non-monetary indicators in the 2016-18 Demographic and Health Survey and estimate the World Bank's Multidimensional Poverty Measure. Despite the significant growth since 2009, monetary poverty and access to several essential services hardly changed, which stands in stark contrast to the substantial improvement across the rest of the world and other comparison regions over the same period. Combined, the results illustrate that it is possible that very little of resource-led growth trickles down to the population and that the link between macroeconomic and microeconomic outcomes is more tenuous in PNG than found in other resource-intensive settings.
with Safa Almoayad, Eliana Favari, Samira Halabi, Siddharth Krishnaswamy, and Almedina Music
Forthcoming in Review of Development Economics
Using a high-frequency survey in Yemen, we demonstrate how school attendance responds to a series of conflict-related shocks. First, there are a number of plausibly exogenous events that significantly change the severity of violence but have limited impacts on school attendance. These events include the capture of the southern capital by secessionist forces, an unexpected partial ceasefire, and the capture of a governorate from the internationally recognized government by Houthi forces. And second, we demonstrate how shocks aside from living in close proximity to violence - institutional declines and macroeconomic shocks associated with the conflict - can have large impacts on school attendance and the ability to afford essential services. Combined the results better illustrate some of the mechanisms by which conflict impedes school attendance and human capital formation and the results have implications for education assistance being delivered in conflict settings.
How Does Violence Force Displacement During Active Conflict? Evidence from Yemen
with Anna D'Souza, Eliana Favari, and Siddharth Krishnaswamy
World Bank Policy Research Working Paper WPS 10177
The ways in which violence forces displacement are not well understood given difficulties in collecting data during conflict. We investigate this issue during Yemen's conflict, which has led to a large forced displacement crisis. First, we demonstrate that violence significantly escalated leading up to and following displacement in the districts from which displaced households fled, and this escalation exceeded that of households that did not become displaced and the regions to which displaced households moved. Second, we demonstrate that the escalation of violence around the time of displacement varied by type of violence. Violence from ground battles escalated leading up to and following displacement- the type of violence with the largest fatalities per violent incident and most associated with the capture of territory; but other prevalent types of violence either peaked prior to displacement or did not appear to be strongly associated with displacement. And third, we demonstrate that there was a significant amount of heterogeneity in the violence experienced by households before displacement. A significant share of displaced households fled during times of no violence, but violence escalated in the regions from which these households fled following displacement. We argue that the last result is likely explained, in part, by these households being more averse to potential violence than other Yemeni households. Combined, these results corroborate that violence is pivotal to forced displacement, but further illustrate the complexities of deciding whether and when to become displaced.
with Tara Vishwanath
World Bank Policy Research Working Paper WPS10179
As conflicts across the globe continue to escalate and data collection in these settings becomes more sensitive, policy makers and researchers are forced to turn to alternative methods to accurately collect vital information. We assess the ability of novel and anonymous internet-based surveys to elicit sensitive information in Yemen's conflict by comparing identical sensitive and non-sensitive questions in an internet survey to a concurrent mobile phone survey. There were significantly larger differences between the modalities in sensitive questions than in non-sensitive questions, with a greater share of respondents expressing sensitive views in the internet survey. Overall, the results are consistent with the internet surveys eliciting more sensitive information than other more commonly used survey modalities, and suggest that internet surveys can be used in conjunction with other techniques to acquire information that would be otherwise difficult to collect.
Drafts Coming Soon:
What are the Most Important Drivers of Food Emergencies During Conflict? Evidence from Yemen
with Anna D'Souza, Eliana Favari and Siddharth Krishnaswamy
Recent research has demonstrated that the onset of conflict is associated with sharp drops in food access, but that following the onset of conflict, the location of violence is not correlated with the prevalence of food insecurity. However, given the lack of high-frequency and regionally disaggregated food security data in most conflicts, we do not have a thorough understanding of how and why food insecurity evolves in these settings. We investigate potential drivers of food emergencies in the midst of active conflict in Yemen using five years of household data from a monthly food security survey of the entire country. During this time period, the country has faced a number of conflict-related shocks, all of which were important to the food security situation. However, we demonstrate that the timing and location of the largest declines in food access were associated with the multiple currency crises that have occurred. Importantly, comparing the impacts of currency crises to other shocks, we demonstrate that rapidly rising food prices in particular were driving the worst food security outcomes, as opposed to increases in the price of fuel or other goods, shocks to income, or the many other shocks that have occurred. Combined, these results demonstrate that rapidly rising food prices are likely an important determinant of food emergencies in conflict settings, and that rapid price rises can be the result of conflict-related factors aside from being in close proximity to violence.
Are Forcibly Displaced Households Always More Vulnerable to Shocks than Host Populations? Evidence from Yemen
with Anna D'Souza, Eliana Favari, and Siddharth Krishnaswamy
Forcibly displaced households often are forced to sell assets, exhaust what little savings they might have, and leave their homes and possessions behind in order to flee conflict. With potentially fewer resources at their disposal to weather shocks, often times forcibly displaced households are more vulnerable to subsequent adverse shocks than host populations. Using five years of monthly household surveys, we demonstrate that forcibly displaced households and host communities in Yemen's conflict were identically impacted by many conflict-related shocks that occurred. Multiple currency crises and a complete air and sea blockade all had nearly identical negative impacts on food access of both groups. Importantly, the similarity in the impacts are invariant to the the time since displacement and a number of pre-shock characteristics of both displaced and non-displaced households. Combined, the results demonstrate that the resilience of forcibly displaced households relative to host populations is context specific, and that there exist conflicts in which host and forcibly displaced populations have similarly exhausted their ability to cope with shocks.