Content uploaded by David Stasavage
Author content
All content in this area was uploaded by David Stasavage on Jun 27, 2014
Content may be subject to copyright.
Why Did Public Debt Originate in Europe?1
David Stasavage
New York University
david.stasavage@nyu.edu
November 2013
1Paper prepared for the volume Fiscal Regimes and the Political Economy of Premodern States,
edited by Andrew Monson and Walter Scheidel.
Abstract
Medieval Europe was in many ways an economic backwater. Why then did such a region
give birth to the distinctly modern innovation of public debt? In this short piece I consider
what political, military, and institutional factors distinguished Europe from other world
regions. I then suggest how these factors in‡uenced the development of public credit.
Introduction
Today we take it as a given that to function e¤ectively, a government needs to have the
ability to borrow, and to borrow over long time horizons. Yet in the long history of …scal
states, public borrowing is a relatively recent innovation. This is a puzzle demanding
explanation. Perhaps even more puzzling is the fact that a generalized form of long-
term public borrowing …rst emerged in Medieval Europe, an economic backwater in many
respects. In this short piece I will suggest why public debt …rst originated in Europe and
what this tells us more generally about the political conditions necessary for a state to gain
access to long-term credit. The initial development of public credit in Europe depended
heavily on a particular political institution –a representative assembly that monitors and
intervenes in the area of state …nance. However, the emergence of representative political
institutions itself depended on a deeper causal factor – the long-standing trend in Europe
for certain cities to be able to govern themselves autonomously. In an era of high travel
and transport costs it was initially only possible to sustain the institutions necessary for
public credit in a small polity, such as an autonomous city. It was only over time that rulers
in Europe’s larger territorial states would learn to establish access to credit, precisely by
working through their cities.
In what follows I will develop my argument in the following sequence of steps. Be-
ginning with an abstract consideration of the factors that condition the development of
public credit, I will then chart the evolution of public borrowing in Europe over a period
of …ve and a half centuries between 1250 and the French Revolution. This will then be
followed by an exploration of why city-states were the pioneers with regard to credit, why
they eventually died out, and …nally how territorial rulers eventually learned to harness
the power of their cities. The account here derives from material in my recent book
(Stasavage, 2011) as well as two recent papers (Stasavage, 2010, 2014).
1
What Conditioned the Development of Public Credit?
If we are to ask why public credit …rst emerged in Europe, then it makes sense to …rst
think in abstract terms about the factors necessary for a government to be able to borrow.
There were four necessary pre-conditions for public credit to emerge.
A …rst and most obvious condition is that a government must have a monetary source
of revenue that can be used to repay any debt that is contracted. Some of the states
considered in this volume had …scal systems, and sometimes even advanced ones, that
were not monetary in form and therefore this condition was not satis…ed.1In Europe,
however, even in the early medieval era where state capacity was weak at best, polities
had access to monetary forms of revenue.
A second condition for public credit to emerge, or more exactly to be desirable, is
that a state must be faced with expenditure shocks. For modern states such expenditure
shocks can come in a number of di¤erent forms involving the need to respond to economic
crisis, sudden needs for infrastructure, or even those arising from natural disasters. For
the polities considered in this volume, in which modern forms of social protection were
absent, wartime needs constituted the almost exclusive form of expenditure shock. Once
we take this into account, we can understand why a full ‡edged public debt did not develop
in a place like Tokugawa Japan. With the problem of civil violence solved and external
threats kept at a safe distance, there were no wartime expenditure shocks and therefore
there was no need to establish a highly developed system of public credit.2The situation
in Europe could not have been more di¤erent. Warfare was prevalent from a very early
date and only increased in intensity as time progressed. It is a matter of debate whether
war was the prime causal factor that drove European state development. What is certain
though is that were it not for the demands of war, European polities would have faced far
fewer needs to borrow.
1For an example, see Terence d’Altroy’s contribution for this volume on the Inka …scal regime.
2For further elaboration see Phillip Brown’s paper for this volume on the Tokugawa …scal regime.
2
Even if a polity is faced with a sudden expenditure shock, there remains an alternative
to borrowing if it is possible to suddenly increase revenue in dramatic fashion. The problem
is that doing this with monetary revenue involves very signi…cant political and economic
costs. However, it is also possible for polities to levy a "tax in kind" for war purposes. The
most common tax of this sort has been military conscription. A number of the polities
in this volume had e¤ective systems of military conscription that avoided the need to
o¤er soldiers monetary compensation for their services and which therefore reduced the
need for developing a system of public credit. Once again, the situation in Europe could
not have been more di¤erent. For European territorial states a system of conscription
did initially exist in the form of feudal service. However, this system evolved towards
one of monetary compensation of paid mercenaries. This development took place for the
simple reason that those bound only by feudal obligation often failed to show up when
summoned. In the autonomous cities of Europe a di¤erent set of circumstances prevailed,
though the end outcome was the same. Initially cities defended themselves and waged war
by drafting citizen militias. Over time, however, it became apparent that a much more
e¢ cient strategy was to hire mercenaries. In both the autonomous cities and territorial
states then, the move away from waging war by obliging citizens or subjects to serve for
free made it necessary to raise money, and often on short notice.
The above three conditions helped dictate why it was necessary and useful for European
polities to have access to public credit. The presence of monetary revenues also explains
why European rulers could repay debts. However, there is of course a fourth critical
condition necessary for a system of public credit to emerge; there must be some expectation
that rulers will want to repay their debts. Given the immediate …nancial advantages of
defaulting, there has to be some constraint on a ruler’s incentive to exercise this option.
One possibility is the fear for one’s reputation. Failure to prioritize debt servicing may
make it di¢ cult or impossible for a ruler to obtain credit from lenders in the future. While
reputational constraints can be binding, it is also well known that in times of crisis, with
3
pressing and immediate needs, the risk of future sanctions may be insu¢ cient to constrain
a ruler.
Given the potential inadequacy of reputation as a constraint, a large body of work
suggests that access to credit will depend above all on having a good set of institutions that
o¤er creditors the assurance that debts will be repaid. In the context of medieval and early
modern Europe it has been suggested that a representative assembly could play this role,
constraining a ruler to repay debt when they might otherwise have preferred not to do so.3
There are two important questions about this however. First, why would a representative
assembly itself prefer debt servicing to default?4Second, if representative assemblies
served this useful function, then why didn’t they emerge in all European polities?
The other attractive feature of the argument emphasizing representative assemblies
is that both long-term credit and political representation were innovations that saw their
most extensive early development in Europe. Along with the other factors listed above, the
relative absence of strong representative institutions in other world regions may provide
further reason why European polities were the …rst to develop an extensive system of public
credit. With this said, even without representative assemblies as they existed in Europe,
we should not ignore the possibility for rulers in other regions to establish alternative
institutional forms that might have helped sustain a system of public credit.
The Rise of European Public Credit in One Picture
As part of the research for my recently published book, States of Credit, I collected a
new data set that charts the development of public credit in Europe from the thirteenth
through the eighteenth centuries. This expands on the prior important work of Stephan
Epstein for his book Freedom and Growth. For a set of 31 European polities my data set
pools together observations from many disparate sources that report the year in which a
3This is an idea of course most closely associated with the seminal piece by North and Weingast (1989).
4This is a point emphasized and considered in Stasavage (2003).
4
Figure 1: Nominal Interest Rates on Long Term Public Debt (sample of 31 European polities)
0 5 10 15
1250 1350 1450 1550 1650 1750
year
Territorial State City-State
5
polity contracted a long-term debt and the nominal interest rate on this debt. In the vast
majority of cases these were actually life or perpetual annuities, as opposed to true debts.
While the Catholic Church in Europe deemed that charging interest on debt was usury,
it did not make the same judgment with regard to annuities for which a regular payment
was made but the “borrower”in e¤ect never returned the principal.
Figure 1 presents the nominal interest on debt for each of the observations in the
data set, distinguishing between two types of polities – autonomous cities and territorial
states. This single picture is in fact very instructive about the evolution of public credit
in Europe over this period of roughly …ve centuries. There is an obvious and immediate
distinction here between the fortunes of city-states and territorial states. If city-states
began issuing long term debt as early as the thirteenth century, at the height of the
medieval commercial revolution, the …rst territorial states did not take this step until
more than two centuries later. This requires an explanation. Likewise, during periods
when city-states and territorial states both issued debt, we can see clearly from the …gure
that city-states were, on average, able to do so on more favorable terms. This too requires
an explanation.
One possible explanation for the above pattern is that rulers of territorial states simply
had no need or no desire to borrow prior to the beginning of the sixteenth century. This
is implausible given that the demands of war placed …nancial strains on monarchs from a
much earlier date. Another possibility is that European monarchs preferred to contract
short-term loans, such as those taken out by Edward III with various Italian bankers,
rather than incur long-term obligations. The problem with this argument is that these
short-term loans almost invariably came at very high rates of interest that re‡ected the
risk involved.
Given the weakness of these the above explanations, the most logical conclusion is that
rulers of large territorial states would have bene…tted from establishing long terms loans
at an earlier date, but they found di¢ culty gaining access to credit. The key question then
6
is what were the underlying factors that di¤erentiated territorial states from city-states in
this regard?
What Was the Secret of Europe’s City-States?
At …rst glance we might think that autonomous cities had an advantage in terms of access
to credit because of their economic structure. These were economic dynamos involving
long distance commerce and proto industry (i.e. textiles), and it is only logical that
there would have been an available pool of lenders in such places. The problem with
this argument is that large territorial states also had numerous, vibrant commercial cities
under their rule that could have supplied a pool of capital. This raises the possibility that
there was something particular about governance in city-states, and this may explain why
they were so successful in obtaining access to credit.
In States of Credit I report the results of an econometric analysis in which it was
asked both what factors were associated with early access to credit, and what factors were
associated with access to credit at lower cost. The results of this analysis present a puzzle.
On the one hand we see clear evidence that the presence of a representative assembly with
strong …nancial prerogatives, and that met frequently, was associated with better access
to credit. However, once we introduce a dummy variable into the analysis to distinguish
between city-states and territorial states we see, not surprisingly, that city-states had
better access to credit, but also that the representative institutions variables lose any
statistical signi…cance. Does this mean that representative institutions simply did not
matter? No, what it suggests instead is that an intensive form of political representation
did favor access to credit, but this type of political representation existed almost exclusively
in the autonomous cities of Europe. Why would this be the case? In what follows I will
suggest that on one level it was the intensive form of political representation within city-
states that was the key to their success in gaining access to credit. However, this form of
7
political representation was itself dependent on two underlying factors: compact geography
and merchant dominance.
Compact Geography and Political Representation
Political representation is often presented as a necessary adaptation to a problem of scale.
If the population of a polity is too numerous for direct democracy, then choose a set of
representatives. If the population of a polity is spread too widely across the terrain, then
adopt the same solution. However, if representation is an adaptation to problems of scale,
ultimately it should be recognized that representative systems are themselves also con-
strained by scale. In medieval and early modern Europe geographic scale posed a major
constraint for any and all who sought to maintain an intensive form of political represen-
tation within their polity. This fact has been cogently argued in a series of publications
by Wim Blockmans (1978, 1998). In an era of high transport and communications costs
in large polities it was costly to send representatives to an assembly. It was also costly for
constituents to then monitor the actions of representatives once they were in place. Under
these conditions autonomous cities had an obvious advantage when compared with their
larger territorial state neighbors. In a recent article I provide econometric support for the
argument …rst made by Wim Blockmans. Based on an original data set on functions and
prerogatives of representative assemblies in a broad set of European states, I …nd very
strong evidence of a negative correlation between geographic scale and the intensity of po-
litical representation. Moreover, this negative correlation continues to hold even when we
exclude the set of city-states from the sample. This rules out the possibility of a spurious
correlation between geography and representation that was driven by some other feature
of city-states that led them to have an intensive form of political representation.
8
The "Virtues" of Merchant Oligarchy
I suggested above that a representative assembly might constrain a government or ruler
to service debt, but it is certainly also plausible that a representative assembly might
express exactly the opposite preference. For example, on several occasions when it met,
the members of France’s Estates General expressed a desire to see the monarchy defer or
default on debt payments so as to avoid what would otherwise be a signi…cant increase
in taxes. Why did the representative assemblies of city-states not follow this same course
of action? The fundamental reason was that in most cases merchants dominated the
membership of these assemblies and of the executive committees that most often ran
them. These were the same individuals who had liquid capital and who purchased public
debt. In sharp contrast, those inhabitants who paid the indirect taxes to service public
debts most often had little lasting political in‡uence. To see the importance of merchant
oligarchy for credit, we can consider what happened when a mercantile oligarchy lost
power, such as in Siena after 1355. When this took place public credit su¤ered.
The pattern of merchant dominance in autonomous cities was itself dependent on
another underlying factor that was speci…c, if not unique, to Europe. As documented by
Wickham (2007) from an early date a pattern was established for most regions of Europe
in which merchants were located in cities and the members of the landed nobility resided in
the countryside. This was a distinctive pattern that set the stage for merchants to become
dominant in the governance of cities in a way that was not always or often replicated in
other world regions.
How City-States Became Obsolete
If urban autonomy was crucial in leading to the development of public credit in Europe,
there nonetheless seems to be a curious twist to the story. As a number of authors have
observed and attempted to explain, over the long run it was the territorial states, and
9
not the city-states, of Europe that became predominant. For some, such as Tilly (1992)
and Bean (1973), this had to do with changing military technology and economies of scale
in war …ghting. In other words city-states died out because they couldn’t raise the large
armies nor a¤ord new and expensive types of forti…cation or weaponry. However, in States
of Credit I argue that city-states maintained a …nancial advantage over their territorial
state neighbors for longer than is often realized, and this allowed a number of them to
serve well after the year 1500AD, a date that is sometimes o¤ered as the beginning of
the era of the territorial state in Europe. So why then did city-states not retain their
preeminence?
The secret of the …nancial success of city-states was that the same merchants who
provided the funds for public credit also controlled all the levers of political power, ensuring
adequate servicing of debt. Merchant oligarchy was therefore good for access to credit.
The problem of merchant oligarchy in city-states was that if it had unambiguously good
implications for credit, the implications of this political system for long run growth were
considerably more mixed. One of the features of an oligarchy in which oligarchs are
themselves engaged in economic activities is that it can provide very secure property rights
for those on the inside of the political system. This can therefore be favorable to economic
growth if those on the inside, so to speak, have access to the most advanced production
technologies. However, a serious problem can then emerge if in order to sustain the
process of economic growth it is necessary to have a continuous in‡ow of new entrepreneurs
bringing new techniques. As long as an oligarchy establishes barriers to entry into markets,
and this was certainly the case for urban oligarchies in medieval and early modern Europe,
then this poses an obvious problem for sustaining economic performance over the long run.
In a recent paper (Stasavage, 2014) I provide evidence consistent with the above theo-
retical mechanism; an oligarchic form of rule was good for city-state economic performance
in the short term but bad for the long term. Using population growth as a proxy for eco-
nomic growth, I show econometrically that during the …rst century in which they were
10
politically independent, autonomous cities in Europe on average grew substantially more
quickly than non-autonomous cities. However, after roughly a century of independence,
autonomous cities stagnated relative to those cities subject to princely domination. This
robust …nding points to a new explanation for why Europe’s city-states eventually became
marginalized despite their excellent access to credit. The same political institutions that
gave them access to credit also condemned them over time to economic obsolescence.
How Territorial Rulers Used the Credit of Their Cities
The initial development of public credit in Europe was highly dependent on the exis-
tence of a particular form of polity, an autonomous city. This same pattern of urban
autonomy would continue to in‡uence the development of public credit when the territo-
rial monarchies of Europe …rst established systems of long-term borrowing. Just as the
autonomous cities of Europe issued annuities for which payment was backed by future
municipal revenues, subject cities within territorial monarchies also often issued similar
…nancial instruments. Initially territorial rulers used this as an ad hoc …nance mechanism
for their own purposes. Subsequently, territorial rulers began to do this more systemati-
cally resulting in the creation of the …rst true national debts. In France this practice was
initiated with the creation of royal rentes sur l’hotel de ville in 1522. Through this system
annuities were issued in the king’s name but were managed by the municipality of Paris,
and the rentes were paid using revenue streams monitored by that same body, or at least
initially so. In Castile a di¤erent system emerged. The monarchy had a representative
assembly that was dominated by eighteen towns. Each of these towns issued annuities
that were then used for royal …nance. The lesson of the French and Spanish experiences is
clear. Lacking the state capacity to have a true uni…ed system of national debt, complete
with the institutions to sustain this, large monarchies found that the optimal solution was
to piggyback o¤ of the forms of governance that had been established by cities within
11
their territories. Though the case of England after 1688 is often, and rightly, hailed as a
very successful system of public credit, England was actually something of an exception
among European territorial states. Compared to other states in this category, it’s core
territory was quite small. A feature that went hand in hand with this is that England’s
…scal system was centralized from a very early date, in fact from the Norman conquest.
Conclusion
In this contribution I have attempted to explore the reasons why a true system of pub-
lic credit …rst emerged in Europe during an era where, paradoxically, the economic and
administrative strength of European polities was considerably weaker than that in many
other world regions. The combination of a monetary economy, sudden expenditure de-
mands due to war, and the lack of a system for mobilizing manpower without payments
made it very useful for rulers of European polities to have access to credit. However,
access to credit also depended on a critical fourth condition, an expectation on the part of
creditors that debts would actually be repaid. It is now well established that one method
for establishing this expectation is to adopt a good set of institutions for managing debt.
Yet this raises a paradox because medieval Europe was a place in which institutions of
governance and state capacity were undoubtedly weak. Drawing on my work published
elsewhere, I have argued that the answer to this puzzle can be providing be investigating
the autonomous city phenomenon in Europe. Europe’s autonomous cities were islands of
state capacity that had the institutions and political composition necessary for the devel-
opment of a system of public credit. Ultimately, however, the same institutions that gave
city-states a …nancial advantage put them at a long run disadvantage and ensured they
would become obsolete.
12
References
[1] Acemoglu, Daron. 2008 "Oligarchic Versus Democratic Societies." Journal of the
European Economic Association. 6:1-44.
[2] Blockmans, Wim. 1978. "A Typology of Representative Institutions in Late Medieval
Europe." Journal of Medieval History. 4:189-215.
[3] Blockmans, Wim. 1998. "Representation (Since the Thirteenth Century)." In The
New Cambridge Medieval History, ed. R. Mckitterick. Cambridge: Cambridge Uni-
versity Press.
[4] Brown, Philip. 2013. "Wages of Peace: Fiscal Perspectives on Early Modern Japan."
paper prepared for this volume.
[5] d’Altroy, Terence. 2013. "Laboring to Explain the Inka Fiscal Regime." paper pre-
pared for this volume.
[6] Epstein, Stephan. 2000. Freedom and Growth: The Rise of States and Markets in
Europe, 1300-1750. London: Routledge.
[7] Grafe, Regina. 2012. Distant Tyranny: Markets, Power, and Backwardness In Spain,
1650-1800. Princeton University Press.
[8] Mokyr, Joel. 1990. The Lever of Riches: Technological Creativity and Economic
Progress. Oxford: Oxford University Press.
[9] Munro, John H. 2003. "The Late-Medieval Origins of the Modern Financial Rev-
olution: Overcoming Impediments from Church and State." International History
Review.
[10] North, Douglass C., and Barry R. Weingast. 1989. "Constitutions and Commitment:
The Evolution of Institutions Governing Public Choice in Seventeenth-Century Eng-
land." The Journal of Economic History 49 (4):803-32.
[11] Stasavage, David. 2003. Public Debt and the Birth of the Democratic State: France
and Great Britain 1688-1789. Cambridge: Cambridge University Press.
[12] Stasavage, David. 2010. "When Distance Mattered: Geographic Scale and the Devel-
opment of European Representative Assemblies.”American Political Science Review.
104:625-643.
[13] Stasavage, David. 2011. States of Credit: Size, Power, and the Development of Euro-
pean Polities. Princeton University Press.
[14] Stasavage, David. 2014. “Was Weber Right? The Role of Urban Autonomy in Eu-
rope’s Rise.”American Political Science Review.
13
[15] Tilly, Charles. 1992. Coercion, Capital, and European States, AD 990-1992, Basil
Blackwell.
[16] Wickham, Chris. 2007. Framing the Early Middle Ages: Europe and the Mediter-
ranean. Oxford: Oxford University Press.
14
