Theory Talk #21: Stephen Krasner

Sunday, October 19, 2008

Stephen Krasner on Sovereignty, Failed States and International Regimes

In International Relations (IR), one commonly refers to ‘states’ without questioning what this word refers to. Stephen D. Krasner has dedicated the last decade investigating ‘weak’ or ‘badly governed’ states, the history of state formation and different elements or forms of sovereignty. In this Talk, he explains, amongst others, why IR is still all about states; why state-building is thus a central challenge; and how one should distinguish between different elements or kinds of sovereignty in order to understand current international politics.


What is, according to you, the biggest challenge or principal debate in current IR? What is your position or answer to this challenge / in this debate?

I think that at this moment the biggest challenge is how to deal with failed and badly governed states, whether it be third-parties interested in intervening or anyone else. The theoretical debate about this issue is amazingly sparse. There’s some discussion, for example, about how natural resource rents undermine effective governance structures. There is a small literature on whether it is more effective for external actors to deploy a limited number of targeted resources, or a larger amount of general purpose resources. There is, however, no well formulated general debate on how to deal with state failure and bad governance. This is strange, since state failure is wrapped up with the biggest question in political science: how do political structures evolve and how do they get to be what they are.

The current perspectives on state development include the following: firstly, there’s the conventional modernization-perspective, stemming from ideas associated with Lipset, which posits that when you have economic development, you get democratization. It doesn’t exactly explain where economic development comes from, but when Lipset wrote his 1959 APSR article, the economic literature was focused on the so-called ‘finance-gap’; if you could close that by providing foreign assistance to these economically underdeveloped countries, you’d have economic development. Alas but the last forty years of economic assistance have proven that this assumption is incorrect. More aid has not led to more growth.

Then there’s the set of arguments about institutionalization from the public-administration perspective, represented by amongst others Francis Fukuyama and Samuel Huntington (especially the latter’s books Political Order in Changing Societies), which holds that you will have political decay if you have political mobilization without political institutionalization. If you look at Iraq or at Gaza, this argument seems quite correct. So for what might be called conventional institutionalization the critical determinant of development is the construction of political institutions – but, this literature says little about where these institutions come from.

A third perspective is rational-choice institutionalism, which says: you get development when people can strike a successful deal, and institutions are there to increase the probability and likelihood of people making deals. Robert H. Bates (Theory Talk #25) is a good example. He just wrote a book called When Things Fell Apart: State Failure in Late-Century Africa (read the introduction here) about Africa, in which he argues that rulers want to stay in power, something they can do, according to Bates, either by conventional modes of taxation or by exploitation. The choice of which path to follow is driven by the circumstances these leaders find themselves in. His argument is that in Africa a major break occurred in the 70s when there was a sharp decline in growth rates as result of the rising energy prices. Declining growth lead to declining tax revenues. Rulers as a response switched from conventional taxation to rent-seeking exploitation, because only rent seeking exploitation could provide them with enough resources to pay off their supporters. The long term consequences for African development were devastating.

The fourth argument focuses on the importance of path dependence. One recent example is a 2005 NBER paper by Acemoglu, Johnson, Robinson, and Yared entitled Income and Democracy (read it here). Path dependence holds that random developments at a particular point in time can place a polity on a certain path or trajectory; once on that path, even if it is not optimal with regard to economic growth or political development, it is difficult to change. Although I am sympathetic to this path dependence perspective, there is certainly no consensus now in the academic literature about whether modernization theory, conventional institutional theory, rational choice institutional theory, or path dependence provides the most useful perspective for understanding the evolution of political systems. Thinking about the utility of thee difference approaches is not only unresolved but also not well structured.

When we are thinking about conventional state-to-state relations we have a rich literature and well defined positions to draw on. For instance, what should the U.S. do about the rise of China? The existing positions are very clear: we should contain China or engage China. These are very well developed positions. Now another question: what should the U.S. do about Afghanistan? The equivalent discussion is much more confused. Here you can see, that there’s a very well-developed structure for thinking about issues of power politics and great powers, but we lack such developed ideas about failed states. Yet the latter set of issues is increasingly the primary pre-occupation of not only American foreign policy but also that of other states.

How did you arrive at where you currently are in IR?

For me personally, the local environment at Harvard when I was a Ph.D student and assistant professor in the late 1960s and early 1970s was very important - I think it’s fair to say that many of the major scholars from my generation in the field were there: Keohane, Katzenstein, Jervis, Ruggie, Nye…, either as students, faculty, or visitors. Keohane and Nye’s work on transnational relations was an agenda setter. I was not sympathetic to their position but it framed much of my early work. Samuel Huntington was my thesis advisor and his thinking on political order influenced much of my subsequent work. I remember when Political Order in Changing Societies was published how much it undermined what was then the conventional wisdom embodied in modernization theory.

Here at Stanford I have been very influenced by rational institutionalism, which focuses on striking deals rather than issues of power. This, of course, is very consistent with liberal institutionalism which has become the most prominent approach to the study of international politics. Then there are the sociological arguments at Stanford that interest me, especially those made by John Meyer, about the gap that can arise between formal institutions and actual behavior. There are these institutions and norms everybody adopts, ritually enacts, and repeats, but which are decoupled from actual behavior. This framework informed the work that I did on sovereignty in the late 1990s.

What would a student need to become a specialist in IR?

I think that our field is so heterogeneous, that the best a student can aim at, is a varied skill-set that allows him or her to choose and pick the right analytical tools for the right problem. So they need to be able to use statistics, case-studies, and historical archives if that’s appropriate. A great example of applying the right tools at the right time is Michael Tomz most recent book Reputation and International Cooperation (read 1st chapter here).

You’ve indicated that the main challenge for IR is state-building. Why is state-building the issue? One would believe the state to be challenged by ‘globalization’, which prompts other forms of governance such as the European Union, international regimes, etc.

The idea of states and state sovereignty evolved in Europe over a long period of time – it was clearly not the only institutional form of political organization, but at least in Europe, states were very successful in displacing other forms of polities in terms of power and economic growth: states displaced empires and city-states, which were the main rivals of the nation-state.

The state form is then copied and pasted everywhere else in the world, even in places where autochthonous development would never lead to anything looking like a state. Now that has lead to a lot of the issues we’re dealing with right now – failing states as phenomena, and state-building as the challenge. And although there is a lot of talk about other forms of governance than the state, I think the key is still state-building, even though I acknowledge that the other candidate, besides the state, is the European Union, in my view a very successful political innovation different from anything we’ve seen in the past, because it’s not a federal state and its not a international organization – it’s authentically sui generis, as they say. The question is, though, could you have the EU replicated in other parts of the world? I’m skeptical, despite the fact that the EU has served as a model for Mercosul in South America and the African Union.

I think it is impossible to replicate the EU, because of three unique phenomena in its history: firstly, the big ‘European civil war’ of the first half of the twentieth century – killing several hundreds of millions of people really made people doubt if what they were doing before worked.

The second unique phenomenon here is the American security umbrella – because of the Cold War, NATO, and plain American interest, security was taken off the table for Europeans. The third phenomenon and also the most important, is that the most powerful European state, Germany, wanted to bury itself in Europe. After the atrocities of the 2nd world war, they wanted to avoid anything that looked like dominating Europe. And that’s something that’s not going to happen in other parts of the world: Brazil, for instance, is not going to want to limit its freedom of action by Mercosul. South Africa and Nigeria will not let Africa curb their national interests.

That’s why I think that, while it would be great for the world, the EU is not replicable – again, simply because of historically contingent and path-dependent factors. The only option you then have left is to go with the least of all evils, and that’s still the nation-state.

So IR is still all about states?

In my view: yes, with the EU as an exception. There have been a lot of discussions about international organizations, transnational companies and NGO’s. The question of course is: how do we understand these as actors in the international environment? They are obviously significant, but these organizational structures operate within a context governed, created and sustained by states. Look at the UN: it’s true there’s some discretion for the Secretary General and the specialized agencies, but the states are there, they have to vote on the budget and the ‘Powerful 5’ of course have a veto. So in order to be effectual, the UN (and other international organizations) has to maintain the support of states. The WTO is clearly an effort to create a self-enforcing regime in the area of trade, but it was based on the agreement of states. This doesn’t mean that states control it. The US, for instance, had to accept mandatory dispute settlement in the WTO to get concessions in other areas that it was concerned about. NGO’s are increasingly important in the international environment, but are operating within a legal context determined by states. The ability of NGO’s to operate in China or Russia has been severely constrained by these countries. I think that’s it’s not just that states are the most important actors in the international system, but they are the constituting elements of it, including of the other entities that move inside that system. But, as the existence of weak states suggests, the system is principally determined by big, powerful states.

In your article Addressing State Failure (2005) you propose three general steps of conflict management in approaching the daunting task of rebuilding weak states after international intervention. Why aren’t Afghanistan and Iraq rebuilt? Did people not follow the steps or don’t the steps hold?

This article was written when I was working for the U.S. government, and thus written from a perspective of what policy makers could do. Since Afghanistan and Iraq pose long and daunting challenges, it might be too early to conclude anything, but I can say this.

First of all, if you think this is a path-dependent process, then it’s not clear how things will evolve; random events can have a major impact on outcomes. What would have happened if Al-Qaida wouldn’t have succeeded in blowing up the Samarra-mosque? You would have probably had a lower level of sectarian violence in 2006, and the surge and associated initiatives might have been even more successful. If you look at the situation in Afghanistan, one of the things you see is that it’s not clear that we could have understood and predicted the situation in Pakistan, which is, of course, highly influential in the conflict in Afghanistan.

Second of all, and this is more of a policy issue, neither the U.S., nor any other developed country, is well equipped for state-building. If you look at the allocation of funding for defense, including for nuclear weapons, we’re talking about 800 billion a year, while the budget of the State Department and the AID combined is 40 billion. That’s a twenty to one ratio. Everybody involved in state-building would agree that that’s not the appropriate ratio, but changing it is really hard. The U.S. doesn’t even have a heavy police unit, some sort of equivalent of the French gendarmerie or the Italian carabinieri. So we don’t have the least kind of capacity needed for post-conflict stabilization.

You wrote a book in 1999 about sovereignty in which you argued that there are different elements of sovereignty. Can you explain?

I distinguish between, firstly, international legal sovereignty, which has to do with whether a state is recognized, whether it gets to be member of international organizations and whether it has the right to diplomatically represent itself. The second is what people generally call ‘Westphalian’ sovereignty, although I think it actually has very little to do with the peace of Westphalia, which encompasses the idea that each state is autonomous; reflected in the rule of non-intervention in the internal affairs of other states. The third is domestic sovereignty, which has to do both with the domestic authority structures and how effective they are.

If you look at Taiwan, it has Westphalian sovereignty, and it has effective domestic sovereignty. But it is not internationally recognized, o more accurately is only represented by a few states, so it doesn’t have international legal sovereignty.

If you look at any member of the European Union, they have international legal sovereignty (they all still have separate representation in the UN, for instance), they have effective domestic sovereignty, but they don’t have Westphalian sovereignty, because they accepted both supernational institutions and qualified majority voting.

If you look at Somalia, it has international legal sovereignty, but it certainly doesn’t have effective domestic sovereignty, and it might or might not have Westphalian sovereignty.

So in effect, there are surprisingly few states that do have all three attributes of sovereignty: if you take out failed and badly governed states and the European Union, and other oddities like Taiwan, you have fewer states with all three kinds of sovereignty than most people would think.

Does the conclusion that states do not necessarily conform with or follow the rules of sovereignty mean that we have to adapt our definition of state or that states are contingent and some other form of organization would be better?

I think that in the contemporary environment, there is nothing that could be better. I think the best we could hope for is a world of what you could call ‘responsible sovereigns’, states that can govern effectively domestically and play by the rules internationally. The only alternative out there is the EU, and, as I indicated before, I don’t think it’s replicable.

You proposed for weak or failed states not to recognize their Westphalian sovereignty and have a combination of indigenous government and transnational governance do the work, a sort of dynamic multi-level governance network. Would you mean to by-pass existing state structures in Africa, looking to embrace perhaps more legitimate forms of locally-rooted governance?

No. The idea of ‘shared sovereignty’ is that weak states, lacking certain capacities, might want to contract out certain aspects of governance; or they might want to bring in external actors and, to speak from a rational-choice perspective, cut a deal.

There are loads of examples of this: I found about 20 or 30 countries that have contracted out their customs services, because the existing services are so corrupt that they prevent political leaders from doing what they want to do. A second example, are the Salomon Islands. The government was falling apart in 2003. Political leaders invited Australia to intervene, and Australia agreed on the condition that the Solomon Islands passed a law inviting them and other actors from the region in. A third example is Guatemala. The government recently contracted with the UN to create an investigative unit with personnel appointed by the UN, to try and investigate major criminal activities in Guatemala. So all these examples show, that countries try to expand the toolbox they have to solve their problems, and one way is by appealing to external actors.

The trick, though, is not to bypass states. And while it is hard, you want to be able to hold states accountable for what outside service providers are doing in their territory, rather than having external actors simply displace state capacity. These outside actors should be a phase of transition, in which the states themselves learn to fulfill the responsibilities that come with sovereignty. You want governments to provide some reasonable set of collective goods for their societies – security, secured property rights, education and health. Even if states contract out these responsibilities, they should still be held accountable. If you look at India, for instance, a great share of the primary education is provided for by private actors, even for poor people. But it’s still the Indian state that provides for the conditions in which that is possible, and which is ultimately responsible for it.

Last question. In 1982, you defined international regimes as ‘implicit or explicit principles, norms, rules and decision-making procedures around which actors’ expectations converge in a given area of international relations’. Since, constructivism has incorporated the study of international norms regimes. Are international regimes as salient as they were when you defined it in 1982, and are their dynamics understandable without social constructivism?

The work on international regimes in ’82 was a collective effort, and I have to admit that at that point I had no idea what this definition implied. If I were to re-write the chapter now I would offer three different definitions of international regimes – the constructivist definition you mentioned; a neoliberal definition, which is: ‘regimes are principles, norms, rules and decision-making procedures that solve market-failure problems’; and there’s a realist definition, which is: ‘regimes are principles, norms and decision-making procedures reflecting the interests of the most powerful states in the system’.


Stephen D. Krasner is Graham H. Stuart Professor of International Relations, Senior Fellow of the Freeman Spogli Institute and Senior Fellow of the Hoover Institution at the University of Stanford. He has been editor for International Organization, has worked at the U.S. Council of Foreign Relations, and has published on issues varying from sovereignty to institutionalism to International Political Economy. Amongst his books, Sovereignty: Organized Hypocrisy (1999) and Addressing State Failure (2005) stand out.

Related links


Print version of this Talk (pdf)