to increase its public debt, the villagers were forced to increase their private debt. In their own words:
Say that they, for a lack of credit, were unable to sell any more annuities as a village community, forcing the villagers to sell 100 guilders worth of annuities, which they mortgaged on their houses and land.45
It turned out that creating public debt, using as a security a community responsibility system to make all inhabitants liable for annuity payments, was impossible for Hilversum. A solution was found in having well-off villagers selling annuities on collateral of their privately owned land. This example makes clear that access to financial markets was not at all evident for villages; it also underscores the pricing mechanisms that ultimately priced Hilversum –and quite a few other villages– out of the market.46
For our understanding of the functioning of financial markets, it would be interesting to know to what degree regional differences existed. When we take a closer look at the premiums paid by towns (figure 3), we see that differences were relatively low in the North and centre of Holland. In the only large town of the South, Dordrecht, premiums show greater diversity, varying from 6,7 % to 8,3 %. Dordrecht’s average interest rate was relatively high at 7,1 %, whereas averages in the North (6,7 %) and centre (6,1 %) were lower.
FIGURE 3
Interest rates on redeemable annuities main towns (1514)
North: area above river IJ, no large towns. Center: area between river oude Maas/ Merwede and river IJ, large towns Haarlem, Delft, Leiden, Gouda, Amsterdam. South: area below river oude Maas/ Merwede, large town Dordrecht.
There were few differences between large and small towns: in the North all towns were small and interest rates did not differ much. In the centre the large towns Haarlem, Delft, Leiden, Amsterdam and Gouda in general paid interest rates ranging from 5,6 % to 6,3 %. They were somewhat better off than the small towns, which in general paid interest rates ranging from 5,3 % to 6,7 % (figure 4). Dordrecht, the only large town in the South, paid relatively high interest rates compared to small towns in this region.
When we take a look at the premiums paid by villages in different regions of Holland, it seems that differences were relatively low in the centre, and somewhat higher in the South (figure 5). In the North we clearly encounter the largest differences. This is the region above the River IJ, which was less urbanized than the other regions; Hoorn (3.600 inhabitants), Alkmaar (2.800) and Enkhuizen (2.300) were the largest towns.47 These were small towns; Holland’s six main towns, which all lay below the River IJ, had about 7.000 to 10.000 inhabitants.
FIGURE 4
Interest rates on redeemable annuities small towns (1514)
North: Alkmaar, Hoorn, Enkhuizen, Medemblik, Edam, Monnikendam, Beverwijk, Naarden, Weesp, Muiden, Purmerend. Center: The Hague, Woerden, Oudewater, Vlaardingen’s Gravenzande, Schiedam, Vlaardingen, Rotterdam. South: Gorkum, Asperen, Heukelum, Arkel, Heusden
FIGURE 5
Interest rates on redeemable annuities 1514 (all villages)
There are several reasons to assume that the low level of integration in the North had much to do with the absense of large towns: the availability of money in large urban economies is likely to have been superior, and wealthy elites in large towns were more likely to invest their savings in financial markets. But it is also possible that this part of Holland was poorly integrated with the economy of the rest of the county.48 Direct evidence for the latter idea comes from the villagers of Petten, in the far north of Holland, who claimed in 1514 not to have sold annuities «because their remote location and poverty reduced their creditworthiness».49
In sum: participants in financial markets negotiated premiums on redeemable annuities ranging from 5 % to 10 %. However, most annuities by far were sold at interest rates around 6 % to 7 %. Interest rates were usually quite similar, in large and small towns and in the countryside. Villages in the North seem to have had most trouble gaining access to capital markets: here we encounter a wide range of premiums paid by villages. The low variation in interest rates seems to indicate that financial markets functioned quite well in the centre and South of Holland: they connected supply and demand in a way that prevented the emergence of «pockets» where premiums were either relatively high or low. The exceptance was the North of Holland, where interest rates show a much greater volatility, presumably due to a low level of urbanization and economic integration with the rest of Holland.
V
At the end of the late middle ages financial markets in Holland brought people looking to invest their savings in contact with public bodies looking for funding. They did so on a local, regional and interregional level. Towns could be particularly successful in attracting savings from foreigners: a considerable share of the public debt Leiden, Groningen and Nijmegen created was contracted with people from out of town, and often even from outside the province. In Holland only small differences in interest rates existed between large towns, small towns and villages: except for the north of Holland there is little evidence for the existence of «pockets» where interest rates were either very high or very low. For the rest financial markets thus appear to have been quite efficient in bringing together supply and demand.
To what degree there was already a certain degree of integration of financial markets is difficult to determine. The absence of pockets in the largest part of Holland might point in this direction, but on the other hand we are unaware of how quickly financial markets reacted to interest rates fluctuations elsewhere.50 Still, it may be useful to dwell a little upon the subject of market integration. First of all, it does not seem unreasonable to ascribe the small range of interest rates we encountered in Holland to the network of brokers towns used to create public debt. These provided towns with information on where to sell annuities abroad. It could well be that these networks helped to reallocate savings in a more-or-less efficient way and thus reduced price volatility. The credit networks of smaller towns and villages may have functioned in the same way, but probably more on a regional than interregional level.
The role polities’ demand played in reallocating savings in late-medieval societies may help to remind us of the importance of political economy in pre-industrial societies. The state often was the main force behind macro-economic phenomena, whereas private initiatives did not yet have a comparable impact.51 Yet, there also was a more intricate and complex relation between public debt and private enterprise, because in order to create interregional public debt, polities depended on community responsibility systems that were tied to trading networks. Interregional public debt was possible because towns made all inhabitants liable for default; this means that their creditors predominantly derived security from the possibilities they would have to apprehend a merchant from the polity they had lend to. Therefore supra-local public debt followed existing trading networks. As a result, financial markets could not exceed trading networks, and possibilities for expansion were limited.52 Or, to return to the towns and villages in the north of Holland: assuming these were indeed poorly integrated in the county’s economy, it is easy to see that this severely reduced their possibilities to attract foreign funding. Our study suggests alternatives existed to the mechanism David Stasavage described. Where demand