Skip to main content

Advertisement

SpringerLink
Log in

A replication note on humanitarian aid and violence

  • Short note
  • Published:
Empirical Economics Aims and scope Submit manuscript

Abstract

A recent systematic review by Zurcher (World Develop 98: 506–522, 2017) finds that humanitarian aid unequivocally increases violence. This note demonstrates that this conclusion is not warranted by a careful examination of the studies included in the abovementioned systematic review. We highlight the incorrect interpretation of econometric results and the lack of robustness in the sampled studies. We also replicate three studies, and report mixed evidence on the effect of humanitarian aid on violence.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Notes

  1. The review also provides a detailed description of the potential causal mechanisms between development aid and violence.

  2. We have reached out to the author to get access to the dataset, but have not received a response.

  3. Note that these papers have been cited, respectively, 56, 46, and 6 times according to Google Scholar (as of 5/14/2019). Zurcher (2017) has been cited 16 times. The moderate amount of citations should not understate the influence of these papers. Their findings are implicitly translated into critical policy guidance (e.g., Central Intelligence Agency’s World Threat Assessment 2014; Global Food Security Act 2016).

  4. The reader is referred to Zurcher (2017) for an in-depth discussion of the links between humanitarian aid and violence. We briefly summarize the key mechanisms in this section.

  5. This applies to Narang (2015) in that some countries do not experience civil war or peace at all during the observed period.

  6. Interestingly, Narang (2015) provides this robustness test in Appendix 3, but results are neither fully reported nor discussed.

  7. We identify a similar problem in Narang (2014). More details can be found in Appendix.

  8. Lindner and McConnell (2018) suggest that it is because matching cannot easily distinguish between short-term fluctuations in outcomes and more structural outcome trends.

  9. Note that the main author in Wood and Sullivan (2015) has confirmed via email that they used this specification in a working paper.

  10. We can compare the coefficients in columns 2 and 3 Table 7, by computing the z statistic following the correction suggested by Paternoster et al. (1998), and suggest that the estimates are not statistically different from one another.

  11. This is not a problem in both DID analyses revisited in this paper.

  12. The interaction term’s hazard ratio is significant at 5% in column 1, suggesting that the effect of aid is different for wars with a decisive victory than the effect of aid in wars ending with a non-decisive victory. However, this result is not statistically robust across Table 2 as the interaction’s significance does not pass the 5% level in other columns (but passes the 10% level).

References

  • Anderson GL, Fleming TR (1995) Model misspecification in proportional hazards regression. Biometrika 82:527–541

    Article  Google Scholar 

  • Asfaw S, Davis B (2018) Can cash transfer programmes promote household resilience? Cross-country evidence from Sub-Saharan Africa. In: Lipper L, McCarthy N, Zilberman D, Asfaw S, Branca G (eds) Climate smart agriculture: natural resource management and policy, vol 52. Springer, Cham

    Google Scholar 

  • Asfaw S, Carraro A, Davis B, Handa S, Seidenfeld D (2017) Cash transfer programmes, weather shocks and household welfare: evidence from a randomized experiment in Zambia. J Develop Effect 9:4

    Article  Google Scholar 

  • Barrett C (2002) Food aid effectiveness: it's the targeting, stupid!. Cornell University Applied Economics and Management Working Paper No. 2002–43

  • Barrett C (2006) Food aid’s intended and unintended consequences, Working Papers 06–05. Agricultural and Development Economics Division of the Food and Agriculture Organization of the United Nations (FAO–ESA).

  • Bellera CA, MacGrogan G, Debled M, de Lara CT, Brouste V, Mathoulin-Pelissier S (2010) Variables with time-varying effects and the Cox model: some statistical concepts illustrated with a prognostic factor study in breast cancer. BMC Med Res Methodol 10:20. https://doi.org/10.1186/1471-2288-10-20

    Article  Google Scholar 

  • Bertrand M, Duflo E, Mullainathan S (2004) How much should we trust differences-in-differences estimates? Quart J Econ 119(1):249–275. https://doi.org/10.1162/003355304772839588

    Article  Google Scholar 

  • Buhaug H (2010) Climate not to blame for African civil wars. PNAS 107:16477–16482

    Article  Google Scholar 

  • Buis M (2010) Stata tip 87: interpretation of interactions in nonlinear models. Stata J 10(2):305–308

    Article  Google Scholar 

  • Burke M, Miguel E, Satyanath S, Dykema JA, Lobell DB (2009) Warming increases the risk of civil war in Africa. PNAS 106(49):20670–20674. https://doi.org/10.1073/pnas.0907998106

    Article  Google Scholar 

  • Chabé-Ferret S (2015) Analysis of the bias of matching and difference-in-difference under alternative earnings and selection processes. J Econ 185(1):110–123

    Article  Google Scholar 

  • Chabe-Ferret S (2017) Should we combine difference in differences with conditioning on pre-treatment outcomes? Working Paper. https://sites.google.com/site/sylvainchabeferret/research

  • Christian P, Barrett C (2017) Revisiting the effect of food aid on conflict: a methodological caution. Policy Research Working Paper 8171, World Bank, August.

  • Christian P, Barrett C (2019) Spurious regressions and panel IV estimation: revisiting the Causes of Conflict. March, Working paper. http://barrett.dyson.cornell.edu/files/papers/Spurious%20Regressions%20and%20Panel%20IV%20Estimation%20paper%20and%20appendix%2015%20March%202019.pdf

  • CIA (2014) World threat assessment of the US intelligence community, senate select committee on intelligence, statement for the record, James Clapper. https://www.dni.gov/files/documents/Intelligence%20Reports/2014%20WWTA%20%20SFR_SSCI_29_Jan.pdf

  • Collier P, Hoeffler A (2004) Greed and grievance in civil war. Oxford Econ Papers 56:563–595

    Article  Google Scholar 

  • Collier P, Hoeffler A (1998) On economic causes of civil war. Oxford Econ Papers 50(4):563–573

    Article  Google Scholar 

  • Couttenier M, Soubeyran R (2014) Drought and civil war in sub-Saharan Africa. Econ J 124:201–244

    Article  Google Scholar 

  • Crost B, Felter J, Johnston P (2014) Aid under fire: development projects and civil conflict. Am Econ Rev 104(6):1833–1856

    Article  Google Scholar 

  • De Ree J, Nillesen E (2009) Aiding violence or peace? The impact of foreign aid on the risk of civil conflict in sub-Saharan Africa. J Develop Econ 88:301–313

    Article  Google Scholar 

  • Dimick JB, Ryan AM (2014) Methods for evaluating changes in healthcare policy: the difference-in-indifference approach. J Am Med Assoc 312(22):2401–2402

    Article  Google Scholar 

  • FAO, IFAD, UNICEF, WFP and WHO (2017) The state of food security and nutrition in the world 2017. Building resilience for peace and food security. FAO, Rome

  • Findley M (2018) Does foreign aid build peace? Ann Rev Polit Sci 21:359–384

    Article  Google Scholar 

  • Ford I, Norrie J, Ahmadi S (1995) Model inconsistency, illustrated by the cox proportional hazards model. Stat Med 14(8):735–746

    Article  Google Scholar 

  • Global Food Security Act of 2016 (2016) US Congress. Public Law 114–195, July 20, 2016. https://www.congress.gov/114/plaws/publ195/PLAW-114publ195.pdf

  • Gupta V, Mason-Sharma A, Lyon ZM, Orav EJ, Jha AK, Kerry VB (2018) Has development assistance for health facilitated the rise of more peaceful societies in sub-Saharan Africa? Glob Public Health. https://doi.org/10.1080/17441692.2018.1449232

    Article  Google Scholar 

  • Hendrix C, Brinkman H (2013) Food insecurity and conflict dynamics: causal linkages and complex feedbacks. Stab Int J Sec Develop 2(2):1–18

    Google Scholar 

  • Hsiang SM, Meng KC (2014) Reconciling disagreement over climate–conflict results in Africa. PNAS 111:2100–2103

    Article  Google Scholar 

  • Hsiang SM, Meng KC, Cane MA (2011) Civil conflicts are associated with the global climate. Nature 476(7361):438–441

    Article  Google Scholar 

  • Hudson J (2013) Promises kept, promises broken? The relationship between aid commitments and disbursements. Rev Develop Finan 3(3):109–120

    Article  Google Scholar 

  • Imbens GW, Wooldridge JM (2009) Recent developments in the econometrics of program evaluation. J Econ Lit 47(1):5–86

    Article  Google Scholar 

  • IPCC (2014) Climate change 2014: synthesis report. In: Core Writing Team, Pachauri RK, Meyer LA (eds) Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland, p 151.

  • Keyzer MA, van Wesenbeeck L (2007) Food aid and governance. In: Bulte E, Ruben R (eds) Development economics between markets and institutions. Mansholt Publication Series, vol. 4. Wageningen Academic Publishers, Wageningen, pp 183–208.

  • Kuriakose AT, Heltberg R, Wiseman W, Costella C, Cipryk R, Cornelius S (2013) Climate-responsive social protection. Develop Policy Rev. https://doi.org/10.1111/dpr.12037

    Article  Google Scholar 

  • Lawlor K, Handa S, Seidenfeld D, The Zambia Cash Transfer Evaluation Team (2015) Cash transfers and climate-resilient development: evidence from Zambia’s child grant programme, innocenti working paper No.2015-03, UNICEF Office of Research, Florence.

  • Lindner S, McConnell JK (2018) Difference-in-differences and matching on outcomes: a tale of two unobservables. Health Serv Outcomes Res Methodol. https://doi.org/10.1007/s10742-018-0189-0

    Article  Google Scholar 

  • MacKenzie TA, Tosteson TD, Morden NE, Stukel TA, O’Malley AJ (2014) Using instrumental variables to estimate a Cox’s proportional hazards regression subject to additive confounding. Health Serv Outcomes Res Methodol 14(1–2):54–68. https://doi.org/10.1007/s10742-014-0117-x

    Article  Google Scholar 

  • Macrae J (2002) The bilaterization of humanitarian response: implications for UNHCR, December, EPAU/2002/15. https://www.unhcr.org/research/RESEARCH/3db7c7114.pdf

  • Magone C, Neuman M, Weissman F (2012) Humanitarian negotiations revealed: the MSF experience. https://www.msf-crash.org/en/publications/humanitarian-negotiations-revealed-msf-experience

  • Marin A, Otto Naess L (2017) Climate change adaptation through humanitarian aid? Promises, perils and potentials of the ‘New Humanitarianism’. IDS Bulletin, [S.l.], v. 48, n. 4. ISSN 1759-5436. https://bulletin.ids.ac.uk/idsbo/article/view/2884

  • Mary S, Saravia-Matus S, Gomez Y, Paloma S (2018) Does nutrition-sensitive aid reduce the prevalence of undernourishment? Food Policy 74:100–116

    Article  Google Scholar 

  • Messer E, Cohen MJ, Marchione T (2000) Conflict: a cause and effect of hunger. Environ Change Sec Rep 7

  • Mora R, Reggio I (2019) Alternative diff-in-diffs estimators with several pretreatment periods. Econ Rev 38(5):465–486

    Article  Google Scholar 

  • Narang N (2014) Humanitarian assistance and the duration of peace after civil war. J Polit 76(2):446–460

    Article  Google Scholar 

  • Narang N (2015) Assisting uncertainty: how humanitarian aid can inadvertently prolong civil war. Int Stud Quart 59:184–195

    Article  Google Scholar 

  • Nunn N, Qian N (2014) US food aid and civil conflict. Am Econ Rev 104(6):1630–1666

    Article  Google Scholar 

  • O’Neill S, Kreif N, Grieve R, Sutton M, Sekhon JS (2016) Estimating causal effects: considering three alternatives to difference-in-differences estimation. Health Serv Outcomes Res Methodol. https://doi.org/10.1007/s10742-016-0146-8

    Article  Google Scholar 

  • Open letter to US congress (2017) http://www.usglc.org/downloads/2017/02/FY18_International_Affairs_Budget_House_Senate.pdf

  • Oxfam (2016) The role of aid in the post-2015 world. Oxfam Briefing Paper. https://www.oxfam.org/en/research/accountability-and-ownership

  • Paternoster R, Brame R, Mazerolle P, Piquero A (1998) Using the correct statistical test for the equality of regression coefficients. Criminology 36:859–866. https://doi.org/10.1111/j.1745-9125.1998.tb01268.x

    Article  Google Scholar 

  • Royston P (2001) Flexible parametric alternatives to the Cox model, and more. Stata J 1(1):1–28

    Article  Google Scholar 

  • Ryan AM, Burgess JF, Dimick JB (2014) Why we should not be indifferent to specification choices for difference-in-differences. Health Serv Res 50(4):1211–1235. https://doi.org/10.1111/1475-6773.12270

    Article  Google Scholar 

  • Skoufias E (2003) Economic crises and natural disasters: coping strategies and policy implications. World Develop 31(7):1087–1102

    Article  Google Scholar 

  • Strandberg E, Lin X, Xu R (2014) Estimation of main effect when covariates have non-proportional hazards. Commun Stat Simul Comput 43(7):1760–1770. https://doi.org/10.1080/03610918.2012.744043

    Article  Google Scholar 

  • Struthers CA, Kalbfleisch JD (1986) Misspecified proportional hazard models. Biometrika 73(2):363–369. https://doi.org/10.1093/biomet/73.2.363

    Article  Google Scholar 

  • Tozier de la Poterie AS, Jjemba WE, Singh R, Coughlan de Perez E, Costella CV, Arrighi J (2018) Understanding the use of 2015–2016 El Niño forecasts in shaping early humanitarian action in Eastern and Southern Africa. Int J Disaster Risk Reduct 30:81–94

    Article  Google Scholar 

  • USAID (2014) (Re)Assessing the relationship between food aid and armed conflict. USAID Technical Brief. October.

  • Wing C, Simon K, Bello-Gomez RA (2018) Designing difference in difference studies: best practices for public health policy research. Ann Rev Public Health 39:453–469

    Article  Google Scholar 

  • Wood RM, Sullivan C (2015) Doing harm by doing good? The negative externalities of humanitarian aid provision of civil conflict. J Politics 77(3):736–748

    Article  Google Scholar 

  • Wood RM, Molfino E (2016) Aiding victims, abetting violence: the influence of humanitarian aid on violence patterns during civil conflict. J Glob Sec Stud 1(3):186–203

    Article  Google Scholar 

  • World Bank (2010) World development report 2011: conflict, security, and development. World Bank, Washington, DC

    Google Scholar 

  • Xu R, O’Quigley J (2000) Estimating average regression effect under non-proportional hazards. Biostatistics 1(4):423–439

    Article  Google Scholar 

  • Zurcher C (2017) What do we (not) know about development aid and violence? A systematic review. World Develop 98:506–522

    Article  Google Scholar 

Download references

Acknowledgments

The author would like to thank the editor and four anonymous referees for useful suggestions that helped improve this paper.

Funding

The author did not receive support from any organization for the submitted work.

Author information

Authors and Affiliations

  1. Department of Economics, Governors State University, 1 University Parkway, University Park, IL, 60484, USA

    Sébastien Mary

Authors
  1. Sébastien Mary
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Sébastien Mary.

Ethics declarations

Conflicts of interest

The author has no conflicts of interest to declare that are relevant to the content of this article.

Appendix

Appendix

1.1 Data sources for replication

Narang (2014): https://www.isanet.org/Publications/ISQ/Replication-Data

Wood and Sullivan (2015): https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/29076

Wood and Molfino (2016): https://www.isanet.org/Publications/JoGSS/Replication-Data

1.2 A discussion of Narang (2014)

Narang (2014) uses Cox and Weibull survival models to estimate the effect of humanitarian aid disbursements on the risk of peace failing (or the duration of peace) between 1989 and 2004. Humanitarian aid is defined as in Narang (2015). His baseline modeling specification ignores the potential endogeneity between humanitarian aid and conflict, and unobserved heterogeneity. We refer the reader to the earlier discussion on this issue. Narang (2014) first finds that humanitarian aid has no effect on the duration of peace. He includes an interaction term between humanitarian aid and a dummy variable capturing whether a war has ended with a decisive victory; the interaction term captures the possibility that the type of war settlements affects the impact of humanitarian aid. In particular, it is hypothesized that if the war ended with a decisive victory, humanitarian aid would be more likely to increase the risk of peace failing. He finds that post-conflict states treated with higher levels of humanitarian assistance exhibit shorter spells of peace, only when conflicts have ended with a decisive victory.

When re-examining Narang (2014), we identify the same issue about the interpretation of the interaction term’s hazard ratio than in Narang (2015), and, as a result, we also argue that the author ignores several results that do not support his conclusions. First, the author correctly interprets the results of Table 1 in that there is no effect of humanitarian aid on the risk of peace failing. However, the author misinterprets the results from model 1 Table 2 when the interaction term (humanitarian aid*decisive victory) is included in the Cox survival model. Narang (2014) wrongly interprets the coefficient associated with the interaction “humanitarian aid*decisive victory” (2.241) as evidence that the risk of peace failing more than doubles for every one-unit increase in the log value of humanitarian aid disbursements.

The hazard ratio for the interaction term should be interpreted as a ratio of hazard ratios. We can calculate the point estimate for the humanitarian aid hazard ratio for the case of a war that ended with a non-decisive victory (2.059), but we cannot provide information on whether the estimate is statistically significant. Given the misinterpretation of the HR in the paper, there is no apparent reason to believe that this estimate is significant. It is noteworthy to mention that this problem equally applies to other columns in Table 2.

Further, the hazard ratios for humanitarian aid are all lesser than one across Table 2, implying that humanitarian aid could lower the risk of peace failing in the case of wars that have ended with a decisive victory, but the effect across columns is statistically insignificant. As the dataset is not available for replication, we cannot provide more precise information about the statistical significance of the aid coefficient, that is, whether the effect is significant at 10%.Footnote 12

Alternately, there is another way to make use of the interaction term’s hazard ratio if one is willing to consider the interaction term’s significance at 10% (indicated by the author in the text). Table 2 in Narang (2014) implies that a war ending with a decisive victory largely decreases the risk of peace failing (as the sign of the hazard ratio associated with a decisive victory is very small and definitely lesser than one), but the positive effect of a decisive victory largely increases as humanitarian aid increases. This result is consistent across Table 2 in his paper. This original result of Narang (2014) does not support the author’s conclusions, but rather provides supportive evidence of the positive (but indirect) impact of humanitarian aid on conflict, in the case of civil wars ending with a decisive victory.

More fundamentally, the author does not provide information about the validity of the PH assumption. As the dataset is not publicly available, we cannot provide further evidence on this matter either. But given the issues highlighted in Narang (2015)’s findings based on similar Cox survival models, we suggest that these results should be taken with much caution. In sum, this discussion highlights the fact that there is little evidence that aid increases the risk of peace failing in Narang (2014). We, however, acknowledge this conclusion is conditional on the limited amount of available information (i.e. no available replication material) (Table 8).

Table 8 Test of equality of coefficients for year FE across treated and untreated areas
Full size table

The null assumption is that a specific year fixed effect for the treated areas is equal to the same year fixed effect for the untreated areas. For example, for the year 1999, in Wood and Sullivan (2015)’s replication, we find that we can reject the null at the 10% level.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mary, S. A replication note on humanitarian aid and violence. Empir Econ 62, 1465–1494 (2022). https://doi.org/10.1007/s00181-021-02064-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00181-021-02064-w

Keywords

JEL Classification

Search

Navigation