Working Papers
Determinants of intended return migration among refugees: A comparison of Syrian refugees in Germany and Turkey
[Revise and Resubmit from International Migration Review]
with Nawras Al Husein
This study assesses whether Syrian refugees intend to return to Syria, taking account of the economic, cultural and institutional differences between their country of origin and the host country. We develop a simple theoretical model on return migration and optimal duration of stay in the host country to identify the potential trade-offs faced by refugees. We then assess the theoretical predictions empirically with a sample of 577 Syrian refugees living in Germany and Turkey. Three return scenarios are considered: (i) ever returning, (ii) returning when it is as safe in Syria as before the war, and (iii) returning within two years. Refugees in the immediately neighbouring country of Turkey are more likely to regard their stay as temporary (76%) compared to those who fled to geographically more distant Germany (55%, p-value of difference=0.000). Concerning the correlates of intended return, we observe that socio-demographic and economic characteristics tend to have limited predictive power for re-migration intentions, independent of the host country. Similarly, while refugees value freedom of speech and belief, the existence of these liberties does not feed into the return migration decision in either of the host countries. Thus, attempts to impose these values on the Assad Government are unlikely to trigger mass return movement. From a policy perspective, we analyse whether random exposure to positive or negative information regarding return migration impacts on the refugees’ intentions. We find no systematic impact on the decision to migrate back. This demonstrates that host governments cannot expect (rapid) information disseminated by refugee agencies – even if it provides support – to impact the refugees’ decision making about return. Overall, the analysis suggests that neither proximate nor distant host countries should bank on the speedy return of the Syrian refugees but should focus on refugee integration, independently of how long they intend to stay.
The impact of off-grid solar home systems in Kenya on energy consumption and expenditures
[Revision submitted to Energy Economics]
with Matthias Rieger, Arjun S. Bedi, Jurgen Vermeulen and Binyam A. Demena
This paper assesses the impact of solar home systems (SHS) on energy consumption and energy-related expenditures among Kenyan households. Based on a pipeline comparison approach of more than a thousand households, we find that access to a SHS leads to a net increase of 24 to 36 minutes in daily lighting use due to a three hour increase in the use of LED lamps, accompanied by a reduction in the use of “dirty” lamps. We also find a one litre reduction in the monthly use of kerosene for lighting. These changes in energy use along with a decline in the cost of charging mobile phones leads to a reduction in overall monthly expenditure of KSh 193 (about EUR 1.60). For the most popular SHS, these economic effects translate into a payback period of about six years. In addition to the economic benefits, we find small positive environmental effects, increased satisfaction from better quality lighting and an increase in time spent watching TV. Setting the costs against the combined benefits of SHS suggests a positive payoff for this particular off-grid electricity option.
Experiential marketing of clean drinking water: Experimental evidence for Kenya and Rwanda
[Under Review with Review of Development Economics]
with Rachel Howell and Kinsuk Mani Sinha
To date limited work investigates why and how consumers in emerging markets make consumption decisions. With the rise in demand for clean drinking water in sub-Sahara Africa, a field experiment was conducted on non-consumers of two socially oriented drinking water companies providing low cost re-usable bottled drinking water in Kenya and Rwanda. Non-consumers were randomly divided into two groups to address whether the provision of purchase-related information and exposure to a physical experience (free sample) increase purchase. For both countries, information motivated purchase (49% for Kenya, 67% for Rwanda) but purchase was even higher among those who received the free sample (67% for Kenya, 78% for Rwanda, p-value<5%). Similarly, planned future purchase and the purchase frequency were significantly increased in the treatment group. The results suggest that similar to consumers in developed markets, emerging market consumers may not only consume on the basis of price, but rather having an experience with a product can minimize purchase risk.
Does Accountability Enhance Service Delivery? Assessment of a Local Scorecard Initiative in Uganda
with Wil Hout and Binyam Afewerk Demena
This article assesses whether the Local Government Council’s Score Card Initiative, implemented in Uganda since 2009, achieved its intended impact of increasing accountability in public spending and service delivery. We analyse a district-level panel dataset (2005-2016) with administrative data, as well as Afrobarometer data on citizen perceptions (2008-2017). Empirically, we exploit the phasing in of the scorecard for a rigorous difference-in-difference framework with district-specific trends. The results show clear but limited measurable impacts of the scorecards on policy making and public services. Scorecard districts appear to spend less of their budgets in comparison with non-scorecard districts. Impacts on service delivery are less pronounced. Further, the scorecard impacts on citizen perceptions of councillor corruption. Our results provide a nuanced quantitative contribution to the literature on accountability by demonstrating that reporting mechanisms about political representatives are not in vain. Like earlier research, we find that impacts also depend on political competitiveness.
Fertility, electricity and television: Is there a link? Evidence from Pakistan, 1990-2012
with Luca Tasciotti and Farooq Sulehria
Pakistan has the second highest fertility rate in South Asia with every woman giving birth to 3.4 children on average. This paper uses three waves of the Demographic Health Survey data to empirically analyse fertility trends in Pakistan between 1990 and 2012; accounting for wealth, the use of contraceptives and birth spacing, this paper looks at three additional pathways for reducing fertility: (i) electrification, (ii) access to TV and (iii) family planning commercials broadcast on television. The results suggest that the direct effect that the access to electricity has on fertility is limited. In contrast, access to television had a significant effect in reducing fertility, especially after the 2000s. To further disentangle the contribution of television to the fertility decline, we also assess the role of family planning commercials. Indeed, in 2006 and 2012, women watching family planning commercials report lower fertility. In our robustness analysis we assess whether radio ownership and family planning messages on the radio have a similar effect on fertility but we can rule out those channels. We provide complementary qualitative evidence on Pakistani soap-operas and argue that the role models and types of households conveyed by the soap-operas are possible causes of the fertility decline.
Determinants of intended return migration among refugees: A comparison of Syrian refugees in Germany and Turkey
[Revise and Resubmit from International Migration Review]
with Nawras Al Husein
This study assesses whether Syrian refugees intend to return to Syria, taking account of the economic, cultural and institutional differences between their country of origin and the host country. We develop a simple theoretical model on return migration and optimal duration of stay in the host country to identify the potential trade-offs faced by refugees. We then assess the theoretical predictions empirically with a sample of 577 Syrian refugees living in Germany and Turkey. Three return scenarios are considered: (i) ever returning, (ii) returning when it is as safe in Syria as before the war, and (iii) returning within two years. Refugees in the immediately neighbouring country of Turkey are more likely to regard their stay as temporary (76%) compared to those who fled to geographically more distant Germany (55%, p-value of difference=0.000). Concerning the correlates of intended return, we observe that socio-demographic and economic characteristics tend to have limited predictive power for re-migration intentions, independent of the host country. Similarly, while refugees value freedom of speech and belief, the existence of these liberties does not feed into the return migration decision in either of the host countries. Thus, attempts to impose these values on the Assad Government are unlikely to trigger mass return movement. From a policy perspective, we analyse whether random exposure to positive or negative information regarding return migration impacts on the refugees’ intentions. We find no systematic impact on the decision to migrate back. This demonstrates that host governments cannot expect (rapid) information disseminated by refugee agencies – even if it provides support – to impact the refugees’ decision making about return. Overall, the analysis suggests that neither proximate nor distant host countries should bank on the speedy return of the Syrian refugees but should focus on refugee integration, independently of how long they intend to stay.
The impact of off-grid solar home systems in Kenya on energy consumption and expenditures
[Revision submitted to Energy Economics]
with Matthias Rieger, Arjun S. Bedi, Jurgen Vermeulen and Binyam A. Demena
This paper assesses the impact of solar home systems (SHS) on energy consumption and energy-related expenditures among Kenyan households. Based on a pipeline comparison approach of more than a thousand households, we find that access to a SHS leads to a net increase of 24 to 36 minutes in daily lighting use due to a three hour increase in the use of LED lamps, accompanied by a reduction in the use of “dirty” lamps. We also find a one litre reduction in the monthly use of kerosene for lighting. These changes in energy use along with a decline in the cost of charging mobile phones leads to a reduction in overall monthly expenditure of KSh 193 (about EUR 1.60). For the most popular SHS, these economic effects translate into a payback period of about six years. In addition to the economic benefits, we find small positive environmental effects, increased satisfaction from better quality lighting and an increase in time spent watching TV. Setting the costs against the combined benefits of SHS suggests a positive payoff for this particular off-grid electricity option.
Experiential marketing of clean drinking water: Experimental evidence for Kenya and Rwanda
[Under Review with Review of Development Economics]
with Rachel Howell and Kinsuk Mani Sinha
To date limited work investigates why and how consumers in emerging markets make consumption decisions. With the rise in demand for clean drinking water in sub-Sahara Africa, a field experiment was conducted on non-consumers of two socially oriented drinking water companies providing low cost re-usable bottled drinking water in Kenya and Rwanda. Non-consumers were randomly divided into two groups to address whether the provision of purchase-related information and exposure to a physical experience (free sample) increase purchase. For both countries, information motivated purchase (49% for Kenya, 67% for Rwanda) but purchase was even higher among those who received the free sample (67% for Kenya, 78% for Rwanda, p-value<5%). Similarly, planned future purchase and the purchase frequency were significantly increased in the treatment group. The results suggest that similar to consumers in developed markets, emerging market consumers may not only consume on the basis of price, but rather having an experience with a product can minimize purchase risk.
Does Accountability Enhance Service Delivery? Assessment of a Local Scorecard Initiative in Uganda
with Wil Hout and Binyam Afewerk Demena
This article assesses whether the Local Government Council’s Score Card Initiative, implemented in Uganda since 2009, achieved its intended impact of increasing accountability in public spending and service delivery. We analyse a district-level panel dataset (2005-2016) with administrative data, as well as Afrobarometer data on citizen perceptions (2008-2017). Empirically, we exploit the phasing in of the scorecard for a rigorous difference-in-difference framework with district-specific trends. The results show clear but limited measurable impacts of the scorecards on policy making and public services. Scorecard districts appear to spend less of their budgets in comparison with non-scorecard districts. Impacts on service delivery are less pronounced. Further, the scorecard impacts on citizen perceptions of councillor corruption. Our results provide a nuanced quantitative contribution to the literature on accountability by demonstrating that reporting mechanisms about political representatives are not in vain. Like earlier research, we find that impacts also depend on political competitiveness.
Fertility, electricity and television: Is there a link? Evidence from Pakistan, 1990-2012
with Luca Tasciotti and Farooq Sulehria
Pakistan has the second highest fertility rate in South Asia with every woman giving birth to 3.4 children on average. This paper uses three waves of the Demographic Health Survey data to empirically analyse fertility trends in Pakistan between 1990 and 2012; accounting for wealth, the use of contraceptives and birth spacing, this paper looks at three additional pathways for reducing fertility: (i) electrification, (ii) access to TV and (iii) family planning commercials broadcast on television. The results suggest that the direct effect that the access to electricity has on fertility is limited. In contrast, access to television had a significant effect in reducing fertility, especially after the 2000s. To further disentangle the contribution of television to the fertility decline, we also assess the role of family planning commercials. Indeed, in 2006 and 2012, women watching family planning commercials report lower fertility. In our robustness analysis we assess whether radio ownership and family planning messages on the radio have a similar effect on fertility but we can rule out those channels. We provide complementary qualitative evidence on Pakistani soap-operas and argue that the role models and types of households conveyed by the soap-operas are possible causes of the fertility decline.
