Communal violence in the Horn of Africa following the 1998 El Niño
|Title:||Communal violence in the Horn of Africa following the 1998 El Niño||Authors:||Weezel, Stijn van||Permanent link:||http://hdl.handle.net/10197/8243||Date:||Dec-2016||Abstract:||This study exploits a shift in Spring precipitation patterns in the Horn of Africa following the 1998 El Niño to examine the effect of climate change on conflict. Using data for Ethiopia and Kenya and focusing on communal conflict the regression analysis links districts that have experienced drier conditions since 1999 relative to 1981-1998 with higher conflict levels. However, the magnitude of the estimated effect is low and the direction of the effect is as likely to be positive as negative. Moreover the results are sensitive to model specification, not robust to using another outcome variable, and do not generalise well to out-of-sample data. The cross-validation illustrates that the model linking droughts with conflict has a relatively poor predictive performance. The results also show that districts with substantial shares of pastoralism experience higher levels of communal violence, something that is well documented in the qualitative literature, but don’t face higher risks following decreases in precipitation levels.||Type of material:||Working Paper||Publisher:||University College Dublin. School of Economics||Start page:||1||End page:||24||Series/Report no.:||UCD Centre for Economic Research Working Paper Series; WP2016/17||Copyright (published version):||2016 the author||Keywords:||Horn of Africa; Climate change; Rainfall; Communal conflict||Language:||en||Status of Item:||Not peer reviewed|
|Appears in Collections:||Economics Working Papers & Policy Papers|
Show full item record
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.