The Challenges of Equitable International Burden-sharing

This article offers a brief sketch of some of the key concepts and dilemmas related to international burden-sharing. It presents the latter as a key aspect of contemporary global governance that nevertheless faces difficult strategic, political and ethical challenges. These often seem to stem from complaints about a lack of equity in burden-sharing and disagreements over how “equity” is to be interpreted. Contrary to this understanding, the article argues that the fundamental problem is rather that as long as a solution cannot be devised to the problems states are trying to manage in cooperation, no equitable sharing of the related burdens can meaningfully occur. This may be trivial to state but is acknowledged rarely compared to how often it might be the case. The article uses various examples to illustrate this, including a detailed look at international cooperation in the face of the Ebola epidemic and against the Islamic State in Iraq and Syria (ISIS). 

The New and the Old: A Tale of Two Eras of Burden-sharing

It is often posited that we live in an era of burden-sharing. There certainly are problems that are – at times rightly, at times wrongly – defined as common challenges that may constitute burdens to be shared, and there are formal as well as informal frameworks of global governance in which some burden-sharing is occurring.

That we may conclude this takes more explanation than what may seem required at first sight, given the seemingly obvious validity of conceiving of “burdens” in a broad sense, with a view to many different issue areas. In fact, the expression “burden-sharing” originates from a specific Cold War/North Atlantic Treaty Organisation (NATO) context and what was primarily implied by it at the time was the collective defence function of the North Atlantic Alliance for the purposes of which comparable, in some sense proportionate, efforts were expected from member states already at the time of the Korean War (Thies, 2009:106-119). Throughout the Alliance’s history – in fact, up to this day – this was measured mostly in terms of countries’ defence spending in proportion to their Gross Domestic Product (GDP) (Zyla, 2015:8).

The use of GDP in this way was always questionable. Nominal and purchasing-power-adjusted GDP values produce different comparisons of individual country contributions. Per capita GDP values reveal different levels of welfare across countries. Moreover, the interpretation of the inequality of defence-spending-to-GDP ratios was never unambiguous, either. Some read the latter as the “exploitation” of the large by the small, having found that comparatively smaller members of alliances tend to spend proportionately less on defence (Olson and Zeckhauser, 1966). Critics of these views have pointed out, however, that (1) large members within alliances can, due to the spending disparity, establish or reinforce their hegemonic position vis-à-vis others and this may translate as a positive result in their security policy; (2) countries with significant defence industry may benefit from increased defence spending whereas those without it face higher expenditure purely as a cost increase; (3) military capabilities are mixed goods that can be consumed for private purposes as well and not only for the collective purposes of one’s alliance – again to the benefit of countries with larger militaries who can use their capabilities in ways actually indifferent to their allies (Sandler and Hartley, 1999:22-57; Cooper and Zycher, 1989:12-13). These are just a few of the objections that arose over time to the exploitation hypothesis. Moreover, as for example Hynek and Marton show, small alliance members often compensate for underspending on defence by doing more in specific missions where this is marginally more important for the alliance (Hynek and Marton, 2011:15-19).

The Post-Cold War world and the inevitable re-focusing on crisis management and other challenges raised further fundamental questions as to the interpretation of what constitutes equity in burden-sharing. Peacekeeping, peace enforcement, international humanitarian aid and development assistance appeared in calculations and the need for a multi-dimensional assessment of country contributions became more self-evident (Cimbala and Forster, 2005; Zyla, 2015:3-4 and 32-33). Some, like Mark Boyer (Boyer, 1989; 1993), have long argued in favour of such a perspective, for example with reference to the need (and the issue of who does what) for financial stability, already during the Cold War. 

An Overview of the Burden-Sharing Experience of Recent Years

Once – as Senator Richard Lugar suggested in the context of the Balkans crises – NATO “had to” go out of area in order to avoid “going out of business,” the realities of the post-11 September 2001 world dictated that its largest undertaking in this respect take place in Afghanistan. Force generation remained a challenge throughout, in spite of there being, in rhetoric, unity over the importance of this mission to the extent that some saw talk of an “elite consensus” across the Alliance as warranted (Kreps, 2010). In reality, governments of the day, especially in multi-party, proportional-representation parliamentary democracies, had to respond to challenges in domestic politics by their opposition related to their commitment to the Afghanistan mission. There was nevertheless a strikingly consistent recurrence of the theme of how combating global terrorism was in the interest of everyone. As Bennett, Lepgold and Unger’s analysis (Bennett, Lepgold and Unger, 1994) of the 1991 Gulf War showed two decades ago – and as, in fact, this can be directly deduced from the premises of the Realist theory of International Relations – when countries look to balance a threat as part of an alliance, it does make a difference to them to what extent they may be threatened compared to others. Just as Portugal was never threatened as much by the Soviet Union or Warsaw Pact forces as West Germany was in the Cold War, it could not plausibly be claimed that each and every country was really evenly threatened by al-Qaida or that consequently each had the same to gain from NATO’s undertaking in Afghanistan. This may explain some of the difficulties of burden-sharing in Afghanistan.

As operations in Afghanistan have already begun to wind down in 2011, the turmoil of what was dubbed the „Arab Spring” shook the Middle East and North Africa, and with leadership from Britain and France and the enabling support of the United States an intervention occurred in the Libyan civil war. The removal of the Gaddafi regime from power was an undertaking formally adopted as its own by NATO, and the leaders of the coalition subsequently expected contributions from every member state. At the end of the day they were disappointed by the less than forthcoming, or in many cases actually absent, support from others. Perhaps it should not have come as a surprise: the case for burden-sharing here was considerably weaker than in Afghanistan. To avoid speaking with the wisdom of hindsight about the actually negative consequences of what happened: the long-term implications of the intervention, and the “day after” in general, were unclear at the time – and there was no direct threat to speak of from the perspective of those from whom a contribution was demanded on the occasion. This was a humanitarian intervention after all, and the imperative of the situation was framed as a moral choice (basically, to prevent a possible genocide in the town of Benghazi), and not as a question of interest.

At the end of 2013, the Ukrainian crisis erupted, resulting in a major increase of tension in the NATO-Russia relationship. The question of what to do divided the Alliance but even in the absence of a fully consensual view of the situation, it was clear even to the more hesitant members that some sort of response was required after the annexation of Crimea by Russia, to send a signal with the aim of deterrence. By this point, the United States was paying over 75% of the cumulative total defence spending of NATO, and loudly and clearly raising this as a problem. Even if the US was not at this point using all of its defence resources in actuality for the Alliance’s purposes (as it never really was, at any point in history), this was a major lack of proportionality and clearly relevant from the point of view of equity. However, it also transpired in the light of the crisis – as well as on the basis of the lessons of the Libya intervention where European coalition participants were in short time running low on precision-guided bombs – that complaints about quantitative issues may be less relevant when even the modernisation of European militaries became largely impossible under the weight of continuing defence spending cuts. It is certainly relevant from a strategic perspective that the Donetsk and Luhansk People’s Republics may have been supplied more tanks by now than the number of tanks Germany and France have. It is, however, even more worrying, with regards to the future, that there is a lack of willingness in Europe to plan militarily for the general possibility of major future armed conflict. For now, the credibility of the NATO response came to a signficant extent from the United States’ willingness to reassure European allies on the Alliance’s eastern periphery.

If we take a look at (1) the number of participating countries, (2) the ratio of resources-committed/spent to resource-demand, (3) the statistical distribution of individual country contributions, and (4) the dominant framework of interpretation as key variables in the background of these and similar collective efforts, measuring the success of mobilisation for burden-sharing, Libya is clearly the most negative example. There, as conditions proved less than ideal, the outside world quickly lost interest in making a significant investment in improving the situation. In Afghanistan’s case the jury is still out, although most observers agree that sufficient force was never generated throughout the mission and that consequently it is no wonder that the present situation is at best an uneasy stalemate with Islamist insurgents. As much as it has been tested so far, the NATO response to the Ukrainian crisis has worked the most convincingly, albeit with a broad distribution of contributions from different NATO member countries and (as the aim is deterrence) against a hypothetical challenge, with the limited aims of reassurance.

This finding may seem counter-intuitive if one thinks of how, out of the three cases, Afghanistan was almost unanimously framed as a common threat in NATO member state leaders’ rhetoric. A simple observation this article seeks to confirm therefore with two additional case studies is that the “framework of interpretation” variable may not really be key to determining the quality of a coalition effort – as measured by “resources-committed to resource-demand” and “distribution of individual country contributions.” What matters more, as will be shown below, is how pure (how purely public) a public good is, and, additionally, how purely it can be produced. 

International Cooperation in the Face of the Ebola Epidemic and the Islamic State

The distinct specific challenges related to the Islamic State and the Ebola epidemic have both been framed as universally concerning by a host of heads of state and government as well as by representatives of international organisations. The efforts against both were consistently presented by almost all state and intergovernmental actors effectively as the production of public goods. The World Health Organisation (WHO) declared the Ebola epidemic in West Africa an “international public health emergency” on 8 August 2014 and received criticism for doing so only belatedly. As far as the Islamic State is concerned, UN Security Council Resolution 2170 concluded that ISIS is responsible for systematic and grave abuses of human rights, including crimes against humanity. Many Middle Eastern countries, among them Jordan and Saudi Arabia, have referred to ISIS as a major terrorist threat by September 2014, one that they thought would have implications for the rest of the world, already before the coming together of the US-led international coalition at the Paris conference on 15 September 2014.

Yet the two challenges are very different. The Ebola epidemic is in effect only an intercontinental flight away from the key central nodes of civic aviation now that it has, over the course of 2014, made its way into urban areas/larger settlements in West Africa. It thus concerns all the nodes of a scale-free global network through the small-world effect of connectivity. In this way, it indirectly threatens even the Islamic State. The latter, on the other hand, does not seem to threaten all corners of the globe in quite the same way. Islamic State-inspired terrorist attacks may be a threat in a large number of countries but they are not a plausible threat in a similarly large number of other countries for the moment. Unlike the Ebola virus which is content with any human host anywhere, the Islamic State’s message cannot similarly effectively make its way everywhere.

The means with which it is possible to act in the face of these challenges are also different. In the case of the Ebola epidemic counter-productivity is rarely an issue in the case of most conceivable measures (albeit some containment/quarantine measures have been criticised in this respect). Certainly, efforts to create a vaccine, isolating patients in Ebola Treatment Units (ETUs), supplying such ETUs to countries with weak public health capacities along with other forms of aid, or the methodical contact-tracing that is required with every new discovered case – such measures generally cannot plausibly backfire. Also, a country that acts against Ebola is typically not facing an increased threat (although health personnel working overseas and becoming sick is a possible vector of importing the disease in the case of medical evacuations). The Islamic State, on the other hand, is very responsive in its propaganda, and joining the coalition against it does evoke at least signals that this might result in an increased threat. It is also unclear if the current measures against the Islamic State may plausibly work: both the effectiveness of air strikes and the different armed forces and groups fighting the Islamic State on the ground can, and have been, questioned.

Not only is thus the action against the Ebola epidemic a clearer case of public good production – it is also clearly more feasible. Joining this action poses no similar capability barrier as having assets useful for air strikes does in the case of combating the Islamic State. International law poses no barrier, either, as it does for instance in the case of bombing targets in Syria without unambiguous consent from the Syrian government there.

From this it can be deduced that in reasonable hypotheses one should expect a greater number of participants, a better resources-committed to resource-demand ratio, smaller distribution of country contributions, formal institutional frameworks of cooperation, and no manipulation of access to the good being produced in the case of cooperation against the Ebola epidemic.

Conforming to this one indeed finds that mobilisation was more successful in the case of the Ebola epidemic. According to the Ebola Funding Tracker database, as of April 2015 sixty-eight countries have contributed some kind of assistance to fighting the spread of the disease. As to combating ISIS, according to a list by the Washington Post, 13 countries have expressed political support, 13 have provided humanitarian aid and 20 have done more than this (for example, they supplied military equipment to forces fighting ISIS or participated in air operations). The State Department has a longer list of sixty countries (plus the Arab League and the European Union) but clearly the majority there is not part of the military action against ISIS.

The results achieved so far, inasmuch as they allow one to draw conclusions as to the resources-provided to resource-demand ratio, also point to the mobilisation against Ebola being more successful. In the case of Ebola contact-tracing has become more efficient, the number of available ETUs and other public health capacities grew beyond needed capacity in some places, and the R0 value (the average number of people infected by a single infected individual) has come down over time, along with the number of new cases. The epidemic has been contained to the three West African countries primarily affected, and of these Liberia already seemed to have ended its epidemic for a brief period of time earlier this year. In contrast with this, ISIS was able to adapt well to the military pressure applied against it and has not lost major territory as a result of air strikes and action on the ground. Demand has been more fully met in the case of Ebola in light of formal projections of anticipated needs by the UN Office for the Coordination of Humanitarian Affairs (OCHA) as well. In September 2014, by which time the WHO officially deemed the crisis “an international public health emergency,” OCHA assessed that funding of $988 million would be needed by the UN – of which a sum of $378 million was committed by mid-October 2014. By November the required funding was re-assessed as amounting to $1.5 billion but pledges and contributions were growing accordingly as well. At the same time significant contributions were made outside the UN system, too, and the Ebola Funding Tracker listed contributions totalling $2.91 billion by state actors alone by April 2015.

A simple measure of difference in terms of the distribution of country contributions may be that in the case of ISIS many coalition participants offered only political support which is hard to price in, whereas in the case of Ebola the sixty-eight countries alluded to above all pledged as well as provided funding or assistance of tangible value.

As to the manipulation of access to the public good in question, in the case of Ebola it would be very hard to imagine how this could be done in the first place. To the contrary when it comes to ISIS: a crucial issue in the organisation of the fight against the Islamic State was how to secure the antagonistic cooperation of distinct coalitions that fear each other’s gains, too – not only those of ISIS. On the one hand, there is Iran, the Syrian government and Hezbollah (along with other, minor armed groups); and on the other there is the U.S.-led coalition including western countries as well as Arab countries, notably from the Gulf region. This explains why bilateral diplomacy, e.g. visits by General Lloyd Austin to the region on behalf of the Pentagon, informal deals with Saudi Arabia, or the Obama administration’s secretly initiated correspondence with Iran over the issue were necessary when preparing for operations against ISIS whereas in the fight against the Ebola epidemic multilateral institutional frameworks dominated, with the WHO taking the lead and others such as the World Bank or specific UN agencies also playing their role. Funding the campaign against Ebola was also organised largely through these channels, complemented by some ad hoc initiatives such as the London conference of 2 October 2014 at the United Kingdom’s initiative. But whereas the UK was a keen financer of the latter, and to some extent a leading country in the fight against Ebola, a coalition leader was much more clearly visible as well as necessary against ISIS. The United States had to fulfill this role even as a key event in mobilising support was seemingly the Paris conference of 15 September 2014 at French initiative – in fact, the latter basically only provided a multi-party seal of approval to the Obama administration’s strategy announced days earlier. 


Given (1) that the dominant framework of interpretation (that burden-sharing is required and legitimate) – which was the same in both the Ebola and ISIS cases – does not effectively predict the quality of burden-sharing, and (2) that in the preceding overview even in a case where there was no consensual framework of interpretation (in the Ukraine case) it was possible to have a working burden-sharing arrangement, one can draw a modest conclusion: namely, that behind dominant frameworks of interpretation strategic narratives may matter more (on the topic of strategic narratives see De Graaf, Dimitriu and Ringsmose, 2015).

There has to be agreement among burden-sharing actors not only that a certain goal is desirable (and that it may be called a public or common good) but also regarding the ways and means of achieving said end. A working strategic narrative has to offer this. It has arguably been missing to some extent in Afghanistan whereas in Libya it was emphasised that a moral choice needed to be made rather than an actual strategy found. To the contrary in the case of Ukraine: although NATO member states may not have agreed completely on a narrative of what happened in Ukraine, on what to offer to Ukraine in the long run or how much they could afford to sanction Russia economically, they all agreed that some kind of signal of military preparedness was needed exactly in order to decrease the ambiguity of the situation after Russia’s annexation of Crimea. That this could be achieved by a comparatively minor build-up of forces along the Alliance’s eastern periphery was a belieavable narrative in the sense of its constituting a set of anticipated cause-effect mechanisms that most could trust.

Similarly, in the case of the Ebola epidemic it was never a questionable narrative that an international mobilisation of resources could, with sufficient effort, achieve results, and thus burden-sharing worked here as well. In contrast, in the fight against the Islamic State the strategic narrative includes many loose threads. What happens in a country such as Iraq with infrastructure ruined even further by air strikes – is it going to offer the prospect of a land free of violent radicalism of the jihadi variety? Does the killing, apparently at a rather slow pace, of Islamic State combatants result in incapacitating the organisation? What happens if that is not the case? And, given continued successful recruitment by ISIS, are Syria and Iraq as battlespaces net killers or breeders of jihadi combatants – for example from the perspective of European countries?

To some extent these findings contradict Mancur Olson’s expectation that collective action may fail with greater likelihood if the good to be produced is a true public good – as is the case with fighting Ebola (Olson, 1975:14-16). Olson convincingly concludes this alluding to how it may be in no one’s interest to act when anyone and everyone may be expected to act. It is not difficult, however, to offer a plausible explanation of the findings here with reference to strategic narratives – one that may function as a caveat to Olson’s proposition. When something can be realistically done and inaction carries significant risk that is more or less evenly distributed, collective action is more likely than in cases where risk is uneven and the good to be produced is either not purely a public good or it cannot be produced purely – that is, without the possible production of some public as well as private “bads.” 


Bennett, A., Lepgold, J. and Unger, D. (1994). Burden-Sharing in the Persian Gulf War. International Organization, 48 (1), p39-75.

Boyer, M. A. (1989). Trading Public Goods in the Western Alliance System. Journal of Conflict Resolution, 33 (4), p700–727.

Boyer, M. A. (1993). International Cooperation and Public Goods: Opportunities for the Western Alliance. Washington, D.C.: Johns Hopkins University Press.

Cimbala, S. J. and Forster, P. K. (2005). The U.S., NATO and Military Burden-Sharing. London and New York: Franc Cass.

Cooper, C. A. and Zycher, B. (1989). Perceptions of NATO Burden-Sharing. Santa Monica, CA: Rand Corporation.

De Graaf, B., Dimitriu, G. and Ringsmose, J. (2015). Strategic Narratives, Public Opinion and War. Winning domestic support for the Afghan War. London and New York: Routledge.

Hynek, N. and Marton, P., eds. (2011). Statebuilding in Afghanistan: Multinational Contributions to Reconstruction. London and New York: Routledge.

Kreps. S. (2010). Elite Consensus as a Determinant of Alliance Cohesion: Why Public Opinion Hardly Matters for NATO-led Operations in Afghanistan. Foreign Policy Analysis, 6 (3), p191-215.

Lugar, R. D. (1993). NATO: Out of Area or Out of Business: A Call for U.S. Leadership to Revive and Redefine the Alliance. Speech to the Overseas Writers’ Club, Washington D.C., 24 June 1993.

Olson, M., Jr. (1975). The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press. (Fifth printing.)

Olson, M., Jr. and Zeckhauser, R. (1966). An Economic Theory of Alliances. Santa Monica, CA: Rand Corporation.

Sandler, T. and Hartley, K. (1999). The Political Economy of NATO: Past, Present, and into the 21st Century. Cambridge: Cambridge University Press.

Thies, W. J. (2009). Why NATO Endures. Cambridge: Cambridge University Press.

Zyla, B. (2015). Sharing the Burden? NATO and Its Second-Tier Powers. Toronto: University of Toronto Press.

Tags: ,

Please Consider Donating

Before you download your free e-book, please consider donating to support open access publishing.

E-IR is an independent non-profit publisher run by an all volunteer team. Your donations allow us to invest in new open access titles and pay our bandwidth bills to ensure we keep our existing titles free to view. Any amount, in any currency, is appreciated. Many thanks!

Donations are voluntary and not required to download the e-book - your link to download is below.