Effective aid in conflict zones

Eli Berman, Joe Felter, Jacob N. Shapiro, Erin Troland, 26 May 2013

a

A

Can foreign aid and development assistance help countries emerge from civil war and unrest? Or does injecting money into conflict zones simply encourage corruption and fuel conflict?

Recent research studying the relationship between food aid and conflict is not encouraging. It shows that increases in US food aid driven by spurts of agricultural productivity in the US lead to increased conflict risk in aid recipient countries (Nunn and Qian 2012). This result is difficult to contextualise since academic research on aid and conflict tends to focus on the role of aid in post-conflict settings. Theoretical models indicate that under fairly broad assumptions the welfare effects of cash aid in places prone to conflict is ambiguous, depending on factors such as how cohesive the society is and the level of government capacity, while technical assistance (if effective) should be conflict-reducing (Besley and Persson 2011).

Aid in ongoing conflicts

While such post-conflict research is important, a significant fraction of aid is now directed at countries currently suffering conflict. The figure below illustrates that between 20% and 40% of global official development assistance has gone to countries in conflict1 with the US and UK contributing slightly more to conflict-affected countries than the European Commission. Though overall levels of civil conflict have declined over the past four decades, aid allocation to conflict zones remains high. This is partially due to good news: secure countries such as China are graduating out of the need for aid. Yet it also reflects a countervailing trend increase in the use of aid as an instrument of international security policy.

Figure 1. Yearly proportion of foreign aid flowing to conflict countries, 1976-2010

Small investment works

When aid is used as a security policy instrument, what kinds of investments can the development community prioritise to prevent the outbreak of civil violence, support households and communities during conflict, and accelerate post-conflict recovery?

The short answer, based on recent research, is this: small projects that lose value to the community when civil violence breaks out or when non-state actors gain substantial sway over territory tend to be violence-reducing. For example, grants which support placing government teachers in schools would qualify, or repeated, small, infrastructure maintenance projects. Building roads and major infrastructure, however, or large-scale investments in skills training do not, and may actually destabilise communities by creating incentives to capture economic rents through violence.

Foreign aid tends to generate corruption

This recommendation is based on four findings:

  • The pervasive evidence that foreign aid may in fact generate corruption rather than reduce it (Svensson 2000, Knack 2001, Olken 2006);

Case studies routinely find that this associated corruption generates popular distrust of host-nation governments and of international institutions, and recent research finds that the graft associated with some kinds of aid may actually increase conflict (Wilder and Gordon 2009, Fishtein and Wilder 2012, SIGIR Report 2013). US and NATO forces’ experiences in Afghanistan provide a current and compelling example of this unintended consequence of foreign aid provision.

Aid spending is not related to growth

  • The consistent lack of strong evidence from cross-national studies that aid spending is systematically related to growth (see e.g. Easterly 2001, Rajan and Subramanian 2008);

This null result could stem from many sources. One likely possibility is presence of two countervailing responses to receiving foreign aid. Aid can help recipient countries escape the self-reinforcing political and economic dynamics that make it hard for such countries to generate economic growth, so-called poverty traps (see e.g. Sachs 2005, Collier 2007). But, such positive effects may be offset by the pernicious incentives aid creates for low investment in governance by recipient countries as well as the challenges of effectively administering development assistance when many parties are tempted to directly enrich themselves with aid flows or have weak accountability mechanisms (see e.g. Ghani, Carnahan, and Lockhart 2005, Easterly 2006, Moyo 2009).

Economic growth and conflict aren’t closely related

  • The link between economic growth and conflict within countries is ambiguous, both theoretically and empirically;

Theoretically, a broad theme in the literature is that rebels will have difficulty mobilising resources, recruits and general support if the public expects rebellion to undermine investments and economic growth. Yet economic growth in locations where people and property are insecure heightens incentives to use violence to capture economic rents (Collier and Hoeffler 2000, Fearon 2008). Empirically, richer countries experience less conflict. Yet within countries the evidence is mixed. Berman et al. (2010), for example, find that changes in employment rates are positively correlated with changes in violence in Afghanistan, Iraq, and the Philippines. Miguel and Satayanath (2004) find that in sub-Saharan Africa major droughts are associated with increased conflict and provide some evidence that this is because droughts reduce economic activity. Dube and Vargas (2013) report evidence consistent with this argument from Colombia, showing that exogenous negative income shocks lead to more violence in coffee producing regions. Yet they also find that positive income shocks lead to more violence in oil producing regions. Moreover, as Blair et al. (2012) show, poor urban Pakistanis are less supportive of militant groups than their fellow citizens, most likely due to their greater exposure to the externalities of violence.

Small-scale service provision is violence-reducing

  • We now have relatively consistent evidence that small-scale service provision by governments is violence-reducing in places with robust policing capacity, either from local forces or due to external intervention, though there are important exceptions;

In research on Iraq we find consistent evidence that small-scale, local aid spending leads to contemporaneous reductions in violence. These effects are even stronger for smaller projects (less than $50K in project cost), where there is a strong government security presence and when some development expertise is available (Berman, Shapiro, and Felter 2011; Berman, Felter, Shapiro, and Troland 2013).2 The figure below illustrates that final point, showing how the presence of a provincial reconstruction team with development experts augments the violence-reducing effect of spending on Commanders Emergency Response Program projects in Iraq.3 More spending on these projects was systematically associated with less violence after controlling for area-specific trends, but was even more violence reducing when local units had access to greater development expertise.

Figure 2. CERP spending

Investing in these kinds of small-scale projects reduces conflict by creating incentives for average citizens to support the government in subtle ways. This is not about swaying those who want to pick up a gun or strap on a bomb, it’s about giving their neighbours a reason to call the local authorities to report suspicious behaviour in an environment in which the government has the capacity to act on that report as well as to protect these cooperative informants from retaliation.4

The evidence recommends both modest objectives and persistent application of painfully acquired insights. The international community has proven relatively ineffective at promoting overall economic growth (Rajan and Subramanian 2008). Even if the evidence linking economic development and conflict were clear-cut, which it is not,5 we have little reason to think that aid can encourage the kinds of growth required to move countries out of the income ranges in which the risks of civil conflict are highest.6 Furthermore, even if the international community could design and implement large-scale aid programmes that would improve economic conditions substantially and avoid corruption (and there is little reason to think that it can), there is no strong evidence to indicate that doing so would reduce conflict.

Conclusions

Taken together, these findings suggest a focus on what we know the aid community can do well: work closely with communities to target small-scale, modest improvements that can be implemented in tough places. If accompanied by a gradual improvement in the quality of governance (through enhanced representation, improved transparency, increased tax collection, service provision, and reduced corruption) current aid recipients can aspire to a long-run improvement in both security and prosperity. Aid programmes should heed the historical record and be careful to enhance rather than undermine those processes. 

References

Bazzi, S, and C Blattman (2011), “Economic shocks and conflict: the (Absence of?) Evidence from commodity prices”, Center for Global Development, Working Paper 274.

Berman, E, M Callen, JH Felter and JN Shapiro (2011), “Do working men rebel? Insurgency and unemployment in Afghanistan, Iraq, and the Philippines”, Journal of Conflict Resolution.

Berman, E, JH Felter, E Kapstein and E Troland (2012), "Predation, Economic Activity and Violence: Evidence from the Philippines", NBER Working Paper 18375.

Berman, E, JN Shapiro and JH Felter, (2011), “Can hearts and minds be bought? The economics of counterinsurgency in Iraq”, Journal of Political Economy.

Blair G, C Fair, N Malhotra and JN Shapiro (2013), “Poverty and support for militant politics: Evidence from Pakistan,” American Journal of Political Science.

Blattman, C and E Miguel (2010), “Civil war”, Journal of Economic Literature, 3-57.

Crost, B, JH Felter and PB Johnston (2013), “Conditional Cash Transfers and Civil Conflict: Experimental Evidence from the Philippines”, Working Paper.

Collier, P (2007), Bottom billion, Blackwell Publishing Ltd.

Collier, P and A Hoeffler (2000), “Greed and Grievance in Civil War,” The World Bank Policy Research Working Paper 2355.

Dalgaard, CJ, H Hansen and F Tarp (2004), “On the empirics of foreign aid and growth”, The Economic Journal.

Dube, O and J Vargas (2013), “Commodity Price Shocks and Civil Conflict: Evidence from Colombia”, The Review of Economic Studies.

Easterly, W (2001), The Elusive Quest for Growth: Economists’ Misadventures in the Tropics, MIT Press.

Easterly, W (2006), The white man's burden: Why the West's efforts to aid the rest have done so much ill and so little good, Penguin.

Fearon, J (2008), “Economic development, insurgency, and civil war”, Institutions and Economic Performance.

Fishstein, P, and A Wilder (2102), "Winning Hearts and Minds? Examining the Relationship between Aid and Security in Afghanistan," Feinstein International Center.

Ghani, A, M Carnahan, and C Lockhart (2005), “Stability, State-Building, and Development Assistance: An Outside Perspective,” Princeton Project on National Security.

Hansen H, and F Tarp (2001), “Aid and growth regressions”, Journal of Development Economics.

Knack, S (2001), "Aid dependence and the quality of governance: cross-country empirical tests," Southern Economic Journal.

Miguel, E, S Satyanath and E Sergenti (2004), “Economic shocks and civil conflict: An instrumental variables approach”, Journal of Political Economy.

Moyo, D (2009), Dead aid: Why aid is not working and how there is a better way for Africa, Farrar, Straus and Giroux.

Nunn, N, and N Qian (2012), “Aiding Conflict: The Impact of US Food Aid on Civil War”, NBER Working Paper 17795.

Olken, B A (2006), “Corruption and the costs of redistribution: Micro evidence from Indonesia”, Journal of Public Economics.

Rajan, RG, and A Subramanian (2008), “Aid and growth: What does the cross-country evidence really show?”, The Review of Economics and Statistics.

Sachs, JD (2005), "Can extreme poverty be eliminated?", Scientific American.

Shapiro, JN and N Weidmann (2012), “Is the phone mightier than the sword? Cell phones and insurgent violence in Iraq”, working paper, Princeton University.

SIGIR (2013), “Learning from Iraq: A Final Report from the Inspector General for Iraq Reconstruction”.

Svensson, J (2000). “Foreign aid and rent-seeking”, Journal of International Economics.

Wilder, A and S Gordon (2009), "Money can’t buy America love," Foreign Policy.


1 Defined as countries experiencing 25 or more battle deaths in a given year according to the excellent Uppsala Conflict Data Program.

2 Though small-scale programmes managed in ways that create opportunities for rebels to use violence to capture or impede aid flows appear to temporarily increase conflict in the Philippines (Crost, Felter, and Johnston 2013).

3 Anecdotal evidence from CERP provision efforts in Afghanistan gathered by the co-authors bolster these empirical findings from Iraq.

4 As Shapiro and Weidmann (2012) show, increasing cellular coverage in Iraq, which made it relatively safer for people to share information with Coalition and Iraqi government forces, reduced violence at both the district level and for specific local coverage areas. This finding is consistent with the claim that the flow of information from civilians to counterinsurgents is a key constraint on the production of violence.

5 The cross-national evidence for a correlation between economic growth and peace is quite strong, but the causality of the links is ambiguous; it is not clear whether countries have less conflict because they are growing or they grow slower because of the potential for conflict which is sometimes realised (Blattman and Miguel 2010). Bazzi and Blattman (2012), for example, find no consistent evidence that commodity price shocks which harm developing countries but are driven by events in other parts of the world predict increased conflict.

6 Civil conflict is rare in democratic regimes, which have historically persisted once a country’s per capita income levels exceed $6055 (Przeworski 2005). Very high estimates of the impact on aid is around a 0.5%-1% increase the GDP per capita for a for 1 unit increase in Aid/GDP (Burnside and Dollar 2000, Hansen et al 2001, Dalgaard et al 2004). But many studies question such estimates, showing that results are quite sensitive to how aid is measured and how the statistical analysis is performed (Easterly 2003). To put these high-end results in perspective, the mean country in these analyses takes in around 5% of its GDP in aid. Average GDP per capita in conflict zones is around 1800 per capita USD. This means it would take the average country experiencing conflict at least 25-50 years at average aid levels to grow into income range for democratic regimes.

Topics: Development, Frontiers of economic research
Tags: aid, civil war

IGCC Research Director for International Security Studies; Professor of Economics, UC San Diego; Research Associate, National Bureau of Economic Research

Co-Director of the Empirical Studies of Conflict Project (ESOC), Stanford University

Assistant Professor of Politics and International Affairs, Princeton University

PhD candidate in economics at the University of California, San Diego

Subscribe