David Stasavage is Professor of Politics at New York University.
[1] States of Credit: Size, Power, and the Development of European Polities was published by Princeton University Press in 2011.
Books and Ideas: In your book, you distinguish between two types of government: city-states and territorial states. How do you define these two types of states? What are your conclusions about their access to credit?
David Stasavage: City-states and territorial states were very different creatures. City-states were cities that had a high degree of political autonomy, sometimes virtually complete political autonomy, with respect to princely rulers. There was actually a continuum of cities: some were completely subject to rulers, some had intermediate levels of autonomy, and some had very high levels of autonomy. What I suggest in the book was that if you compare city-states and territorial states, then you observe that they tended to have very different types of representative institutions. City-states tended to have a more intensive form of political representation. We might think that this more intensive form of political representation allowed them to gain access to credit because creditors were represented in these institutions. They could monitor what was going on with respect to credit and debt and so on.
That’s the first part of the argument, but if you dig a little deeper, then you observe that these institutions could only exist in certain places. Here I relied on work by the Belgian historian Wim Blockmans [2] who suggested that at this time in European history an active representative assembly could only exist in a geographically compact polity. This could be the size of a city or something larger than a city but certainly not as large as something on the scale of France, for example. On the first level, then, you might conclude that it was good institutions that provided access to credit. But if you look a little deeper what you see is that the type of institutions that you could sustain depended upon whether you were large or small.
City-states, Political Representation, and Credit
Books and Ideas: According to the standard interpretation, city-states began borrowing earlier than larger principalities because they were more economically developed. Why is that argument unsatisfactory?
David Stasavage: There are three reasons to suggest why simple economic circumstances do not suffice and why we have to resort to political factors. The first is that territorial states also had rich cities within them. The second is that city-states were not always more prosperous. They were often more prosperous to start off with but then fell into periods of economic stagnation. Yet they continued to borrow at low rates of interest. The third reason, which I think is critical, is that political change within a given city had consequences for that city’s access to credit. On the whole these cities were governed by merchant oligarchies, but there were periods when these oligarchies lost power. What I show in the book is that during these periods there was less favorable access to credit. That shows that within the same city, with the same economy, when politics changed so did access to credit.
Books and Ideas: How did you come up with the hypothesis of a link between representative assemblies and public debt?
David Stasavage: I am not the first one to come up with this hypothesis about representative assemblies. For at least a century people have associated the presence of representative assemblies with the ability of states to raise funds. The idea was that if rulers ask for consent then it becomes easier to raise revenue through taxation. A representative assembly might administer taxes or monitor what was done with those taxes. That pertains to revenue. As far as public debt goes, a new line of thinking began with a famous article by Douglass North and Barry Weingast from 1989 about the Glorious Revolution and its role in promoting access to public credit in England. [3] So the idea of a connection between assemblies and finance has been around for a while. But I think it is not enough to ask whether the presence of an assembly matters. What we need to ask is what type of assembly made a difference? And why did that type of assembly exist in certain places and not others? I think that is one of the innovations of my book.
Books and Ideas: Throughout the book you emphasize a link between “intensive forms of political representation” and access to public credit. What do you mean by this term? Is it a euphemism for merchant oligarchy?
David Stasavage: By intensive what I mean is an assembly that meets frequently and because it meets frequently is able to sustain certain prerogatives that would not otherwise be possible. Such an assembly could monitor and modify spending decisions in a way that an assembly that met say once every three years could not feasibly do. The term is meant to express how frequently an assembly met and what prerogatives it could sustain. It does not refer to how broad representation was, nor does it refer to one particular social group being more prominently represented than others. So it is not a euphemism for merchant oligarchy.
However, in Europe at this time, mercantile dominance and intensive representation tended to go together. Merchants were dominant in city-states, which were small enough to sustain an intensive form of representation. So the two tended to go together, but there were exceptions. As I mention in the book, there were some small, territorial states that had somewhat intensive forms of representative as well as large, territorial states like France that had very non-intensive forms of representation.
Why Size Matters
Books and Ideas: In explaining access to credit, you stress not only how often an assembly met and who participated, but also the geographic size of the polity. Is it possible to disentangle these different factors?
David Stasavage: One way is to look at episodes within city-states when the merchants lost power and see what happened with respect to access to credit. When they lost power because of a revolution of the craft guilds or other groups that did not want to pay more taxes to reimburse public debt (these city-states had very regressive tax systems), access to credit would suddenly become more difficult. That shows that it was not enough to have intensive representation. There had to be intensive representation with mercantile dominance.
Showing the effects of geographic scale is much more difficult, because these polities did not change much in size over time. Ideally we would have information for all of these polities about costs of travel and communication. I gathered the available evidence for an article I published in the American Political Science Review, [4] and I summarized this in my book. The data reveal a strong correlation between having an intensive form of political representation and having a geographically compact polity. There is evidence to show that as states became less compact, they were less likely to have an intensive form of political representation.
Books and Ideas: You use the phrase “access to credit” by which, we assume, you mean the ability to borrow at more or less favorable interest rates. But did all states actually want to establish long-term debt?
David Stasavage: It was participation in warfare that led states to want to borrow. Initially European states did not pay soldiers (which would require debt) but relied upon other arrangements. In territorial states there were feudal obligations. Autonomous cities had militias in which everyone had the obligation to serve. Feudal obligations turned out to be ineffective in getting knights to show up and so they had to be paid, which required borrowing. Likewise in the cities they shifted to hired mercenaries, which generated the need to borrow. As would be true today, when governments needed to suddenly, massively, and temporarily increase funding for some project, the logical choice was to borrow in order to smooth out the effect of the increased taxation over time.
Books and Ideas: Would not larger states have sought to borrow more money and would not this have made it more difficult to convince creditors to loan to them?
David Stasavage: The book provides evidence on when states began to borrow and the rates of interest at which they borrowed. It does not provide evidence on the amounts that they borrowed. That is one point where it is weak, but there is a very good reason for that: given the documentation we have, it is a lot easier to learn that (for example) Castile in 1572 was able to borrow at x rate of interest. It is a lot harder to come up with estimates of the aggregate stock of debt held by these polities. It could be possible that city-states borrowed less and territorial states borrowed more, and that drove interest rates up in territorial rates. But I do not think that this was the case because the partial data that we have does not show a tremendous difference in debt relative to revenues for the territorial states that borrowed the most and the city states that borrowed the most.
Not One but Two Types of States
Books and Ideas: What are the implications of your results for our understanding of state formation in early modern Europe?
David Stasavage: The usual story suggests that war was the driving force of state formation and that over time changes in military technology gave advantages to larger states, and larger states won out over smaller states, in particular independent cities. My book suggests that the effect of war may have been much more ambiguous than is often believed. There were two types of states. Territorial states had large populations and could mobilize a lot of people, but they were not very good at mobilizing money. City-states were small, but they had good access to credit and so they were good at mobilizing money. I think it is important to recognize that these two different types of states, each of which had strengths and weaknesses, coexisted in Europe for much longer than is often realized. That is the first conclusion.
The second conclusion is about the importance of geographic scale in determining the type of political institutions that were created. In fact geographic scale may have been as important or more important than warfare in generating the type of institutions that early modern states had. What we need is a more nuanced view than the standard argument that war drove state formation.
Books and Ideas: Your study also attempts to explain the decline of city-states. Could you briefly explain how your hypothesis differs from the usual explanation of this decline?
David Stasavage: The usual explanation for the decline of city-states is that they were simply taken over militarily, and I think that was often the case. But there’s another aspect of their decline that is less often referred to. In initial periods, city-states often had high rates of economic growth and then their economies became stagnant over time. Why was that the case? That’s a question I am working on now, and so I do not provide any evidence in the book, but I do speculate about the following possibility: city-states were dominated by merchant oligarchies, and these oligarchies were political systems that were closed to new entrants (bringing new technologies, for example). These oligarchies were very good for raising finance because the same people who loaned to government controlled government, but it may be that they became stagnant over time because they were not politically open enough; they were too resistant to new entrants. That may explain why a lot of city-states survived as independent cities for a very long time but lost their economic prominence. As I say in the book, they became “rentier-republics.”
Using Corporate finance and Game Theory to Look at Historical Data
Books and Ideas: Could you describe some of the methodological challenges you faced in attempting to create a coherent data set covering so many states over such a long period?
David Stasavage: The debt data in the book are essentially an expansion and a revision of earlier work done by Stephan Epstein for his book, Freedom and Growth. [5] That was quite a fascinating effort to compile data on interest rates at which different states and cities borrowed over time. I reinvestigated all of his sources and tried to find as many other sources for new polities and for new periods of time as I could. I standardized everything and tried to provide information about what types of loans these were and so on. That was the first part of the work.
The second part of the work was compiling information on representative institutions. This was much more difficult because I had to go to different secondary sources and try to infer what types of assemblies existed in the different polities at different points in time. I decided to ask four questions about each polity: (1) Was there an assembly of any sort at all? (2) If there was an assembly, did it have the right to veto taxation? (3) Did it have some role in administering taxation? (4) And finally, the prerogative that is the most important for access to credit: if there was an assembly, did it have some rights to monitor and modify expenditure decisions?
I had to consult all of the available sources to see if for a given polity in a given period these prerogatives existed or not. For some of these states, we know these things quite well; for others, we do not know them at all. For this reason there were cases for which the data collection process was very certain, and there were cases for which the data collection process was more like educated guesswork. What I try to emphasize in the book is that while this dataset on representative institutions is necessarily imperfect, it is also more extensive than what was previously available.
Books and Ideas: In chapter four you apply models of corporate finance and game theory to this data. Could you describe your motivations for doing so? Is there any risk of anachronism in applying modern concepts such as “control rights” to understand the behavior of people living in the 13th and 14th centuries?
David Stasavage: I resorted to modern theories of corporate finance because I think it can always be useful to imagine what was going on in more abstract terms in order to better understand the argument that we are making. The point of using game theory is not to suggest that the world is exact and certain. The point is to try to present what I think was going on in an abstract fashion and thereby better understand it.
There is always a risk of anachronism. These people were living in a completely different time. They did not have the benefit of reading (for example) Jean Tirole’s corporate finance textbook. [6] But I think all the evidence shows that people understood the basic principle that if you were lending to a state and you had a seat on the city council then you would have some influence over what was going on and could increase the likelihood that you would be repaid. People understood this fact and that is why seats on the city councils were hotly contested. In other words, they understood the principle of what we would now call a control right.
Books and Ideas: To test your hypothesis, you used both statistical analysis and case studies. Did you observe discrepancies in the results? Did the case studies provide examples that helped you to develop and refine the statistical analysis?
David Stasavage: People often present qualitative and quantitative work in opposition but I think that it can be useful to combine the two and that was precisely the goal of this book. The case studies were meant to do several things. First of all, statistical tests and data about a broad set of polities can show correlations but they cannot reveal the mechanisms of what was going on. It is only by digging into the cases that I can confirm whether the inferences I’ve made from the statistical tests actually make sense or not. Also, in doing the case studies I realized certain things that helped me to refine the statistical tests. For example, I observed instances where merchant oligarchies lost power within a city-state. Then I returned to the statistical evidence and tried to determine if it was generally the case, statistically speaking, that when merchant oligarchies lost power access to credit suffered.
Books and Ideas: What new research questions emerged for you as a result of this book?
David Stasavage: This is a study of debt rather than a study of economic development or economic growth. An idea that I mentioned in the introduction and in the conclusion—that being a city-state with political autonomy may have been good for growth in the short run but bad for growth in the long run—is worthy of further research and I am currently writing a paper on that subject. The other topic worth pursuing is how geographic scale and access to information conditions the format of political representation. I think it would be important to consider the introduction of new technologies, such as the railroad and the telegraph, to see how they influenced the behavior of representative assemblies. Could representative assemblies meet more frequently and do things that they could not do previously? Were constituencies linked to representatives in a more direct way? Much of the discussion of the public sphere comes from the English case, but England was a very compact polity compared to many other states, and so we might expect certain developments to occur in England before other places where distance was more of an obstacle.
