Version classiqueVersion mobile

State Cash Resources and State Building in Europe 13th-18th century

 | 
Katia Béguin
, 
Anne L. Murphy

Markets for Public Debt Securities

Revisiting Notes on the History of the Interest Rate: Bond Markets in the Republic of Genoa

David Chilosi

Texte intégral

Introduction

1In 1952, the late Carlo Maria Cipolla published an article destined to be among the most widely read of a highly distinguished scholarship. Its main title translates as ‘Notes on the history of the interest rate’. There, Cipolla traced the development of interest rates paid on the shares of the Casa di San Giorgio, the consortium of creditors responsible for the management of the Genoese public debt in the sixteenth century. The figures have subsequently become the standard source for analysing the cost of public borrowing in the Genoese republic, which at the time was one of the most important financial centres in Europe.

2There are two good reasons that make it worth revisiting them. Firstly, in spite of their popularity, reservations have been expressed about the reliability of the figures, both by Cipolla himself and by subsequent commentators. Secondly, if valid, the figures would offer a unique insight into the relationship between official and market rates across the transition from a forced loans system into an open market for bonds, but their significance in this respect has so far been overlooked.

I. Cipolla’s and Later Warnings

  • 1 C. M. Cipolla, ‘Note sulla storia del saggio d’interesse. Corso, dividendi e sconto dei dividendi d (...)
  • 2 C. Cuneo, Memorie sull’antico debito pubblico: mutui, compere e Banca di S. Giorgio in Genova, Geno (...)

3Cipolla1 warned that the author of one of his main sources, Cuneo,2 ‘can only be trusted so much’. Yet Cipolla found that Cuneo’s data matched closely those found in a late eighteenth-century anonymous manuscript held in the university library of Genoa; a few discrepancies could be easily dismissed as Cuneo’s or his publisher’s transcription errors. Neither of these two sources specified the basis of their calculations, either in terms of what sources were used or how the yearly means of the secondary market prices were computed. However, in a few selected years, figures from notarial and administrative documents held in Genoa’s State Archive also matched closely those reproduced by Cuneo and the anonymous writer. Armed with this corroborating evidence, Cipolla concluded that the data could be considered reliable, subject to two caveats.

4The first caveat is that a significant portion of the shares of the debt (luoghi di San Giorgio) was formed of so-called moltiplichi. For example, in 1597 these accounted for c. 28% of the debt. The moltiplichi were shares inherited by religious institutions under the conditions that: they could not be sold on the secondary market; the dividends on these shares had to be re-invested in bonds until a pre-determined sum had been reached; and this sum had to have a specific philanthropic use. This implied inelasticity of demand of San Giorgio’s shares with respect to the rate of return, which was therefore kept artificially low, to a varying extent.

5In part to address this issue, Cipolla not only computed the market rates of the Genoese bonds, but also the yearly market rate implied by the discounts applied to the dividends (proventi or paghe). In fact, if the creditor wanted to cash in the dividend at the time it was declared, instead of waiting until maturity, this was discounted according to the value of the so-called lira di paga. To compute this second series, Cipolla assumed that the dividends were paid with a delay of four-and-a-half years after they were declared.

6He labelled the two series A and B, respectively. As it turned out, they broadly, but only broadly, corresponded: thus, the rate implied by the discount on the proventi was consistently and significantly higher than the secondary rate on the luoghi. Evidently, Cipolla speculated, either the effect of the moltiplichi may have been significant, or, differently from the rate of discount on the dividends, the rate paid on the luoghi may have been depressed by adverse economic conditions. Still, some patterns, he highlighted, were robust; notably, both A and B displayed a sudden fall in the cost of capital from 1570.

  • 3 C. M. Cipolla, ‘Note sulla storia…’, op. cit., p. 260.

7The second caveat that Cipolla emphasized is that the relationship between the money of account and the actual money used the make payments was unstable. However, calculations found in the 1597 report of the doge Senarega made him confident that in the sixteenth century there was no significant difference between the lire di numerato, which was used by the Casa as the money of account to record the value of the luoghi, and the lire di paga, which was used to pay the dividends. Cipolla used this as grounds to focus on the period 1509–1625 only. As Cipolla put it: ‘I am convinced that for whole of the 16th century there was no difference in value between the lire di paga (at the end of the five years) and the lira di numerato, which affected the patterns of A and of B’.3

  • 4 F. Braudel, The Mediterranean and the Mediterranean world in the age of Philip II, vol. II, 5th ed. (...)
  • 5 S. R. Epstein, Freedom and growth. The rise of states and markets in Europe 1300–1750, London, Rout (...)
  • 6 D. Stasavage, States of credit: size, power and the development of European polities, Princeton, Pr (...)
  • 7 S. Homer and R. E. Sylla, A history of interest rates, 4th ed., Hoboken, Wiley, 2005.

8In spite of Cipolla’s warnings, Braudel4 went on to publish the figures, without feeling it necessary to comment on possible sources of error. Perhaps mindful of potential complications with the two series, other scholars have tended to focus on one or the other. Thus, whereas Epstein5 reports means computed on the basis of series B, Stasavage6 looks at series A, while stressing that series from earlier periods probably used a different method, and thus caution is recommended. Although Homer and Sylla7 drew on both series, they also focused on series A, by reproducing it in almost its entirety. However, they comment, the rates are astonishingly low; pending further research, they cannot be considered as representative of the cost of capital at the time.

  • 8 M. Fratianni, ‘Government debt, reputation and creditors’ protections: the tale of San Giorgio’, Re (...)
  • 9 J. Heers, Gênes au XVsiècle, Paris, SEVPEN, 1961, p. 170.
  • 10 L. Pezzolo, ‘Bonds and government debt in Italian city-states 1250–1650’, in N. Goetzmann and K. G. (...)

9At least implicitly, Fratianni8 offers a reason to focus on A. Cipolla’s assumption that the delay with which the dividends were paid was constant throughout the sixteenth century had been challenged by Heers9 and, in a private conversation, by Felloni, for whom Cipolla’s series B suffers from a positive bias. However, Fratianni does not clarify why it is safe to neglect changing ratios between money of account and actual money, when he extends Cipolla’s series to 1739, using the rates published by Cuneo (1842). Pezzolo10 also extends Cipolla’s series A using Cuneo’s data, as well as other sources, but he employs yet another approach: he concentrates on official returns, implicitly assuming that since these are not affected by the idiosyncrasies of the market, they are best suited to be compared with those that obtained in other places.

II. Reliability

  • 11 H. Sieveking, ‘Studio sulle finanze genovesi nel Medioevo e in particolare sulla Casa di San Giorgi (...)
  • 12 A. Lobero, Memorie storiche della Banca di S. Giorgio di Genova, Genoa, Tipografia Ponthenier, 1832 (...)

10Between the seventeenth and the first half of the eighteenth century, there was a progressive worsening in the quality of circulating money, with the consequence that the market equivalent of the money of account steadily increased, by a non-negligible margin: thus, the market value of the scudo of account in actual lire rose by 69% between 1602 and 1675, increasing further over the following decades.11 At the same time, the value of the lire di numerato remained fixed in terms of scudi of account at the 1602 market value of the latter.12

  • 13 D. Peri, Il negotiante, book II, Venice, Presso Gio, Giacomo Hertz, 1682, p. 96; A. Lobero, Memorie (...)
  • 14 A. Lobero, Memorie storiche…, op. cit., p. 161.

11However, Cipolla was simply wrong in assuming that a different denomination was used for the value of the shares and that of the dividends. Other authors, in fact, are unanimous in asserting that the paghe, when they reached maturity, like the luoghi, were denominated in lire di numerato.13 As Lobero14 put it: ‘if the share-holders waited … they would cash the whole return of twenty soldi di numerato’. Hence, while it is not surprising that Cipolla found no difference between the values of the paghe at maturity and the lire di numerato for the sixteenth century, his worries that this relationship changed in the following period were unjustified.

  • 15 D. Peri, Il negotiante, op. cit., p. 96.
  • 16 C. Cuneo, Memorie sull’antico debito pubblico…, op. cit., pp. 121–127.
  • 17 Ibid., pp. 308–311.

12It is likely that Cipolla’s confusion on this issue arose out of Peri’s15 (rather bizarre) claim that there was a difference between the denominations of the paghe before and at the time of maturity, which implied opportunities for arbitrage. Such a view was nevertheless contested by Cuneo,16 for whom both the paghe and the luoghi were denominated in lire di numerato: ‘Many have claimed that in San Giorgio the moneta di paghe was different from that of numerato, but in fact there were only lire di numerato… the moneta di paghe was in essence the value of the paghe written in money of numerato’. As it happens, in spite of Cipolla’s diffidence, Cuneo’s interpretation is the more plausible of the two. Even more than during the period examined by Cipolla, the lire di paga remained remarkably stable between 1626 and 1764, instead of significantly rising at the same time as the quality of the circulating coinage was steadily eroding: the series is trendless and its coefficient of variation declines from c. 8% to only c. 3%.17 It is therefore legitimate to conclude that series B, which is based on the values of the lire di paga only, is not directly affected by the afore-mentioned changes in the market value of the money of account.

  • 18 Ibid., p. 127.

13As for series A, this depends not only on the dividends and their discounts, but also on the secondary market price of the luoghi. According to Cuneo,18 the latter tended to rise as a result of increases in the premium commanded by the lire di numerato over the actual lire, as well as responding to the changing reputation of the Casa. However, this should not be taken as meaning that secondary price of the luoghi was denominated in actual lire. In sharp contrast to what this hypothesis would predict, the secondary price of San Giorgio’s shares almost continuously declined from 1603 until the later seventeenth century, to remain overall stable thereafter, despite the occasional temporary fall. In turn, this suggests that as the intrinsic value of the luoghi was sheltered from changes in the quality of the coinage, their secondary prices were affected only indirectly by them. Thus, what seems to have happened is that the 1602 regulation fixing the value of the lira di numerato in terms of scudi of account succeeded in enhancing the reputation of San Giorgio’s shares: their secondary prices jumped up from 155% of par to 224% of par between 1602 and 1603 to return to the 1602 level only in the 1680s. In short, it is safe temporally to extend series A, too.

  • 19 G. Felloni, Inventario dell’ Archivio, 2013, http://www.lacasadisangiorgio.it/main.php?do=node&tag= (...)

14Thanks to Giuseppe Felloni,19 we now know that the paghe of 1518 were paid between 1521 and 1522, the delay increased up to a maximum of 116 months for the dividends of 1553, and declined to 64 months by 1582, to stabilize itself at around that level until the mid- eighteenth century. Felloni has also kindly provided this author with precise delays for dividends declared in 1518 to 1641. Figure 1 uses Cipolla’s figures until 1625 and Cuneo’s figures for 1626 to 1641; it compares series A and B assuming a constant delay of four-and-a-half years and the variable delays found by Felloni. Following the original notation, the latter series is labelled as ‘B adjusted’.

Figure 1. Yields on San Giorgio’s shares and on San Giorgio’s dividends with constant and variable delays, 1522–1751

Figure 1. Yields on San Giorgio’s shares and on San Giorgio’s dividends with constant and variable delays, 1522–1751

Notes: A = secondary market returns on San Giorgio’s shares; B = yearly rates of discount on San Giorgio’s dividends with constant delays; B adjusted = yearly discounts on San Giorgio’s dividends with variable delays.

Sources: C. M. Cipolla, ‘Note sulla storia del saggio d’interesse. Corso, dividendi e sconto dei dividendi del Banco di S. Giorgio nel secolo XVI’, Economia internazionale, 5, 1952, p. 270; C. Cuneo, Memorie sopra l’antico debito pubblico, mutui, compere e Banca di S. Giorgio in Genova, Genoa, Stamperia dei Sordomuti, 1842, pp309–311; private exchange with G. Felloni in 2013.

15Beginning with the period 1522–1625 examined by Cipolla, after adjusting the figures for variable delays, the average difference between A and B drops from –0.975 percentage points to only 0.014 percentage points; the hypothesis that the two series have the same mean can no longer be rejected. Evidently, the depressing effect of the moltiplichi on the secondary market yields of San Giorgio’s shares was not significant and the match between the two series was tighter than Cipolla originally thought. This is, of course, reassuring.

16Nonetheless, clearly, the story changes in the intervening period. The means between the two series were significantly different after 1625. Regardless of whether one adjusts the figures, there is an obvious and persistent gap between A and B, on average –0.851 or –1.135. The spread suddenly increased in the mid-1620s; henceforth, series B became consistently higher than series A by such an extent that it is difficult to blame the observation on the effect of the moltiplichi alone. We shall return later to the reasons underlying these different relationships between the two series before and after 1625.

  • 20 C. Cuneo, Memorie sull’antico debito pubblico…, op. cit., p. 121.
  • 21 D. Peri, Il negotiante, op. cit., p. 96.
  • 22 C. Desimoni, ‘Tavole dei valori’, op. cit., p. 411.

17Here it is sufficient to notice that Cipolla’s assumption that series B was based on a mean of secondary market discounts, rather than on the primary market discount, is inexact. In fact, Cuneo20 is explicit in stating that the dividends were discounted by the Casa: Ufficio delle paghe: ‘waiting for the time of maturity [the paghe] would have been paid 20 soldi; but on the same day as they were declared, they were bought from the Office at 16.8 soldi’. Peri21 also stated that ‘the price … of each lira di paga was … established by the Magistrate’; and so did Desimoni: ‘the Casa had the privilege that … its dividends were mature only after four years; however those creditors who presented themselves at other times were paid with a proportional discount’.22

18After 1625, the Genoese republic ‘came of age’ and began to issue its own annuities, like the Monte S. Bernardo perpetuities, which it directly managed, independently from the Casa di San Giorgio. If Cuneo’s, and by implication Cipolla’s, figures were reliable, one would expect to find a close match between the rates paid on the two securities, given that they were both perpetuities traded on the same market. Figure 2 compares the official interest rates paid on San Giorgio’s shares with those paid on annuities issued by the republic.

Figure 2. Official interest rates on San Giorgio’s shares (present value) and the republic of Genoa’s debt (yearly minima) and primary yields on San Giorgio’s dividends, 1625–1751

Figure 2. Official interest rates on San Giorgio’s shares (present value) and the republic of Genoa’s debt (yearly minima) and primary yields on San Giorgio’s dividends, 1625–1751

Notes: B adjusted = primary yields on San Giorgio’s dividends.

Sources: San Gorgio’s shares: C. Cuneo, Memorie sopra l’antico debito pubblico, mutui, compere e Banca di S. Giorgio in Genova, Gênes, Stamperia dei Sordomuti, 1842, p. 309–311; republican debt: D. Chilosi, ‘Risky institutions: political regimes and the cost of public borrowing in early modern Italy’, working paper 177/13, London School of Economics, Economic History Department, 2013.

19Clearly, there is a close match between official returns on San Giorgio’s shares and the republican debt, even if the latter tended to be somewhat lower. The hypothesis that their means are the same is accepted; the difference between the two averages is a mere 0.049 percentage points. That the rate tended to be somewhat higher on San Giorgio’s shares than that on the republican debt is to be expected, given that variable returns implied risk. That the difference was nevertheless very small also agrees with expectations, given that the return on the luoghi di San Giorgio was remarkably stable during this period; its coefficient of variation was only about 13%.

20Turning to B adjusted, this series tended to be lower than the other two: the differences between the means are –0.285 with San Giorgio’s official returns and –0.235 with those of the republic; both figures are statistically significant. This is to be expected given that B adjusted refers to short-term, rather than long-term, lending, unlike the other two series. Again, that the difference was nevertheless small is expected. The premium paid for borrowing over a longer term should decline with the levels of liquidity of the market and rise with the risk of conversion or default. All these factors predict a low premium in early modern Genoa, which had a highly developed secondary market, witnessed hardly any significant conversion from the 1660s onwards, and was known for the reliability of its public debt.

III. A Tale of Two Markets

Figure 3. Secondary market prices of San Giorgio’s shares (yearly means, in % of par)

Figure 3. Secondary market prices of San Giorgio’s shares (yearly means, in % of par)

Sources: C. M. Cipolla, ‘Note sulla storia del saggio d’interesse. Corso, dividendi e sconto dei dividendi del Banco di S. Giorgio nel secolo XVI’, Economia internazionale, 5, 1952, p. 270; C. Cuneo, Memorie sopra l’antico debito pubblico, mutui, compere e Banca di S. Giorgio in Genova, Genoa, Stamperia dei Sordomuti, 1842, pp309–311; J. Heers, Gênes au XVsiècle, Paris, SEVPEN, 1961, p. 630.

  • 23 H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1906, p. 89.

21As seen in figure 3, while secondary prices tended to be much lower than par until the sixteenth century, the reverse was true in the intervening period. That Italian forced loans tended to be traded on the secondary market at below par, by a significant margin, is well-known. What is less often noticed is that, while most Italian states abandoned forced loans from the 1520s, these remained an important source of finance in Genoa during the sixteenth century. For instance, major forced loans were imposed on the local nobility in 1565 and 1624.23

  • 24 J. D. Tracy, A financial revolution in the Habsburg Netherlands. Renten and renteniers in the count (...)
  • 25 L. Pezzolo, ‘Bonds and government debt in Italian city-states, 1250–1650’, in N. Goetzmann and K. G (...)
  • 26 See in particular M. Fratianni, ‘Government debt…’, op. cit.
  • 27 L. Pezzolo, ‘Bonds and government debt…’, op. cit.

22The usual explanation for the particularly low secondary prices of forced loans is that they were risky assets. Theoretically, default on voluntary purchases was less likely than on forced loans, as it was bound to jeopardize future sales.24 Empirically, both late medieval Florence and Venice defaulted on their forced loans, to an increasing extent.25 Though sound, the argument is nevertheless partial; in particular, this view sits uneasily with the widely held belief that creditors’ management of the public debt implied that the Casa di San Giorgio was a ‘safe pair of hands’: save for rare exceptions, due to unusual financial strains, its creditors could expect to receive their full dividends at the time they were told they would.26 To be sure, in the fifteenth century, the secondary prices of San Giorgio’s shares were not as low as those of local bonds in Florence and Venice.27 However, the fact remains that the luoghi were traded on the secondary market at levels which were nowhere near par, both in the fifteenth and for most of the sixteenth centuries.

  • 28 H. Sieveking, ‘Studio sulle finanze genovesi nel Medioevo e in particolare sulla Casa di San Giorgi (...)
  • 29 H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1906, pp. 39–41, 97 and 99; J. Heers, Gên (...)
  • 30 G. Giacchero, Il Seicento e le compere di S. Giorgio, Genoa, Sagep éditrice, 1979, p. 257.

23The solution of this puzzle is that, as stressed by Sieveking,28 in late medieval Genoa the luoghi were only ideal units and significant discounts were usually applied on primary issues, too. The examples that he provides suggest that primary yields were somewhat lower than secondary ones, but they nevertheless tended to be closer to the latter than to the official rates: forced loans issued in 1412, 1418 and 1419 commanded discounts of 80 to 90%; in 1435, 220 luoghi were sold at 73 to 78% of par; in 1444, the shares were issued at 65 to 76% of par, while they were traded at an average of 55% of par on the secondary market; in 1464, when the average secondary market price was about 33% of par, the shares were sold on the primary market at 50% of par.29 As noted elsewhere, the forced loan of 1624 was also issued at a rate close to the prevailing secondary market price at the time: 260% of par, as compared to 278% of par on the secondary market.30

  • 31 D. Peri, Il negotiante, op. cit., p. 97.

24The only direct evidence that republican bonds tended to be traded at par on Genoa’s primary market is Peri’s31 assertion that one of these luoghi had been bought for ‘one-hundred scuti’ on the primary market and sold for 105 on the secondary market. Nevertheless, we do find evidence of a break in 1625. As implied earlier, before that date, the yearly yield implied by the discounting of the paghe on the primary market was close to the secondary market yield, and mostly higher than the interest rate; in the intervening period, by contrast, the short-term primary yield became close to the interest rate, and significantly higher than the secondary yield.

  • 32 H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1905 and 1906; A. Molho, ‘The state and p (...)

25This break suggests incentives to conceal the effective rate of interest paid on forced loans. Since in the Italian city-states the servicing of the debt was mainly financed with indirect taxation of primary consumption, it had highly regressive redistributive effects, which periodically led to popular revolts.32 Hence, the political economy of forced loans implied that the city-states’ rulers had every interest in concealing the actual weight of the debt on the popular purse. This was compounded by the fact that, naturally, the lack of a primary market for the bonds implied little transparency.

  • 33 H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1906, p. 27; S. Homer and R. E. Sylla, A (...)
  • 34 S. Homer and R. E. Sylla, A history of interest rates…, op. cit.; J. H. Munro, ‘The medieval origin (...)

26Moreover, loans to the city-state had to be portrayed as a purchase to be canonically accepted. As the trope was at odds with the coercive nature of forced loans, these were represented as a tax or a gift, and, unless very low rates of interest were charged, forced loans were regarded as usurious.33 By contrast, the canonists were more tolerant towards voluntarily bought annuities; in relation to them, the expectation was that the price had to be ‘just’; only exorbitant rates were considered usurious.34

  • 35 L. Pezzolo, ‘Bonds and government debt…’, op. cit.
  • 36 Sources: per capita weight of the debt in Florence and Venice: L. Pezzolo, ‘Government bonds and cr (...)
  • 37 J. H. Munro, ‘The medieval origin…’, op. cit., pp. 516–517.

27Consistent with these arguments, across Genoa, Florence and Venice, we detect particularly low secondary market prices and official interest rates after the mid-fourteenth century, but much less so before.35 The second half of the fourteenth century also saw sudden and significant expansion in the size of their debts, implying negative shocks on real wages: between c. 1350 and 1400, the per capita weight of the debt, as measured in kg of silver per inhabitant of the city, grew by 189% in Genoa, 750% in Venice and 1300% in Florence.36 At the same time, in strict correlation with the consolidation of the Florentine debt, the canonical legitimacy of forced loans came under scrutiny.37

  • 38 L. Pezzolo, ‘Bonds and government debt…’, op. cit., and ‘Government bonds and credit markets…’, op. (...)

28Conversely, across the same three cities, the movement to an open market was accompanied by sudden increases in the official rate of interest, but not in the size of their debts.38 Hence, unless significant discounts were regularly applied to primary issues of forced loans, it is difficult to understand how voluntary markets became so much more sustainable for their urban finances.

IV. The Republican Mark-Up

  • 39 F. Piola Caselli, ‘The formation of fiscal states…’, op. cit., p. 292.
  • 40 D. Chilosi, ‘Risky institutions…’, op. cit., pp. 10–11.

29That from the beginning of the seventeenth century, the secondary prices of San Giorgio’s shares tended to be much higher than par suggests that primary buyers of Genoese bonds systematically enjoyed a profit, which could be cashed in by selling them on the secondary market. This is what happened in the papacy where, typically, new annuity issues were sold at below par to bankers, who sold them at above par on the secondary market.39 Above par secondary prices were unusual for early modern annuities.40 Hence, reasoning that this feature in Genoa and the papacy had a similar cause is not far-fetched.

  • 41 G. Felloni, Gli investimenti finanziari genovesi in Europa tra il Seicento e la Restaurazione, Mila (...)

30There is also other corroborating evidence. Between the later sixteenth and the early seventeenth centuries, the Genoese began to heavily invest in Rome, Milan and especially Venice. By 1673 they owned as much as a third of the debt of the republic of Venice, more than the local patricians. In around 1720, their share of the papal and Milanese consolidated debts was probably in the region of 15%.41 As shown in table 1, in spite of financial integration, the spread between Genoa and Milan, Rome and Venice, rapidly increased during the seventeenth century, instead of converging towards the levels implied by differences in risk. This suggests the Genoese market was not competitive and that falling secondary prices signalled declining profits for primary buyers, which offset falls in transactions costs.

Table 1. Bai–Perron test results

Spread

Break date

Fitted

Ratio

Yearly rate of change (%)

Beginning

Maximum

End

Before

After

R-squared

N

Milan/Genoa

1711

1.611

2.902

2.335

0.692***

0.307**

0.475

69

(6.402)

(2.204)

Rome/Genoa

1675

1.126

1.840

1.624

1.045***

0.261***

0.751

41

(9.262)

(5.053)

Venice/Genoa

1650

1.433

2.453

1.584

1.482**

–0.347***

0.444

37

 

 

 

 

 

(2.746)

(–3.998)

 

 

Notes: The tests allow both the constant and the yearly rate of change to vary after the break date; two break dates are allowed; the figures in parentheses are t-statistics.

Sources: D. Chilosi, ‘Risky institutions: political regimes and the cost of public borrowing in early modern Italy’, working paper 177/13, London School of Economics, Economic History Department, 2013.

  • 42 M. Flandreau and J. H. Flores, ‘Bonds and brands: foundations of sovereign debt markets’, Journal o (...)

31In modern financial parlance, systematic differences between primary and secondary prices of assets are called initial public offering discounts. These were a regular feature of nineteenth-century European bond markets and reflect asymmetric information about risk between investors. Discounts are expected to increase both with the risk of the asset and the extent to which the primary buyers enjoy monopolistic power, as this implies a rent or mark-up. Following on from this logic, Flandreau and Flores42 have developed a model that predicts that few prestigious under-writers with monopolistic power tend to focus on low-yield, safe bonds, and enjoy high profits on these. By contrast, primary markets for high-yield, risky bonds tend to be competitive, and therefore produce low profits.

  • 43 G. Felloni, Gli investimenti…, op. cit., pp. 90–91.

32There are clear analogies between this model and the early modern Italian annuity market. The distinction between prestigious and common primary buyers echoes that drawn by Felloni43 between merchants and bankers. The former type of intermediary was prevalent in economically backward places with little financial development and few links with Genoa. The latter, by contrast, were prevalent in financially developed cities where the Genoese frequently invested; they were often from Genoa. In short, it seems that the Genoese bankers had a high market share of bond under-writing and their role at the time can be seen as being akin to that played by the Rothschilds in nineteenth-century Europe: their involvement signalled safe assets.

  • 44 D. Chilosi, ‘Risky institutions…’, op. cit.

33Consistent with this perspective, both Genoa and the papacy, where we find evidence that primary buyers enjoyed significant mark-ups, can be characterized as issuers of safe assets, with comparatively low interest rates. To some extent, such features were shared by Venice. As argued elsewhere, if as usually held, low interest rates in Genoa were rooted in its oligarchic republican constitution, the Venetian spreads with Genoa tended to be in excess of what one would expect on the basis of the riskiness of Venetian bonds.44

  • 45 G. Felloni, Scritti di storia economica…, op. cit., p. 176.

34There remain, nonetheless, difficulties with the model. The extent to which in Genoa the local bankers played an intermediary role analogous to that played in foreign markets is unclear. Traditionally, there, the oligarchs were forced to buy shares on the primary market. Moreover, the moltiplichi, as well as the code di redenzione through which the republic redeemed the debt, implied inelasticity of the demand on the secondary market for bonds, and thus increased the scope for mark-ups for primary buyers at the expense of philanthropic institutions and taxpayers. Accordingly, by the end of the eighteenth century, an unusually high share of the local debt was owned by Genoese religious and philanthropic institutions (c. 69%), rather than private investors (c. 21%).45 Evidently, while there were only a few small private investors benefiting from the primary buyers’ signalling, the oligarchs’ vested interests were well-served by strong ties with the rulers.

  • 46 D. Chilosi, ‘Risky institutions…’, op. cit.
  • 47 G. Felloni, Gli investimenti…, op. cit.; K. Béguin, Financer la guerre au XVIIsiècle. La dette pu (...)

35In this latter respect, Genoa was similar to Rome, where at least since the times of the Medici, the papacy cultivated close ties with the big Italian banking houses, using public funds to pursue policies of patronage. Moreover, not all places where the Genoese invested can be characterized as safe for ordinary investors. Thus in Milan, both the risk of default and interest rates were high by Italian standards. Payments to ordinary investors were regularly repudiated by Milan through frequent partial defaults; only important investors with strong ties to the local ruling class could expect good conditions of lending.46 Milan was not the only exception that confirmed the rule of the sound judgement of the Genoese. Similar risky conditions for ordinary investors prevailed in France, another favourite destination for Genoese capital.47

  • 48 G. Felloni, Scritti di storia economica, op. cit., p. 176; L. Pezzolo, ‘Bonds and government debt…’ (...)

36Consistently above par secondary prices in Genoa suggest a stationary risk: it looks as if Genoese bonds were considered safe throughout the seventeenth and eighteenth centuries. At the same time, Genoese bonds were overwhelmingly locally held: the residents owned as much as 92% of San Giorgio’s shares in 1629; it is likely that the proportion remained stable at around this level until the end of the eighteenth century.48 Hence, it is difficult to account for the fall in the mark-up detected in the decades around the mid-seventeenth century in terms of risk or competition.

37These dynamics can be explained by the political economy of the republican mark-up. Republican rulers, just like any other, faced an incentive to minimize the cost of public borrowing. The fiscal sustainability of the debt demanded balancing the books and new taxes were unpopular. At the same time, unlike the ruling classes of principalities, they were directly accountable to the local oligarchs. Because of this, not only, as often stressed, did they have incentives to minimize the risk of default on the public debt, but also to maximize the return enjoyed by the local oligarchs. In short, they faced contradictory incentives; which one prevailed at which time depended on the relative weight of interests.

38Thus, the greater the share of the Genoese oligarchs’ financial investments was in the local bond market, the stronger the incentive for local rulers to offer advantageous conditions to them on the primary market. This is consistent with particularly high mark-ups in the 1620s, when, in the wake of defaults by Philip II, Philip III and Philip IV in 1596, 1607 and 1627, the Genoese capitalists had severed their links with the Spanish crown and brought their capital back home. In the intervening decades, at the same time as the Genoese began to heavily invest in Venice and other foreign bond markets, however, it looks as if the mark-up fell.

  • 49 G. Felloni, Gli investimenti…, op. cit.

39By the same token, as the Venetian patricians gave up being the majority owner of the local debt in favour of the Genoese over the course of the seventeenth century, both the incentive for rulers to offer mark-ups to primary buyers and the credibility of their commitment to repay the debt weakened. This logic can account for the comparatively rapid fall in the spreads with Genoa after the apex was reached in the mid-seventeenth century. By the eighteenth century Venetian bonds were traded on the secondary market at par or below, signalling low or no mark-ups, and delays of payments and partial defaults increasingly put off the Genoese from further investing there.49

Conclusion

40In summary, Cipolla’s figures emerge pretty well from our ‘grilling’. Cipolla’s concerns about the relationship between money of account and effective money turns out to be a ‘red herring’; adjusting for variable delays implies a much closer match between Cipolla’s series than he originally thought, for his period; and, in the subsequent period, we observe a close match between official rates paid on the republican debt, on the one hand, and on those paid on San Giorgio’s shares and the primary yields implied by the discounting of San Giorgio’s dividends, on the other. We were also able to account for a few discrepancies between these three series.

41Although direct evidence is scant, there are both theoretical and empirical grounds to believe that secondary market yields overstated somewhat the actual cost of public borrowing for Genoa when it relied on forced loans. This is as expected: otherwise coercion would have been redundant. Nevertheless, the primary yields of forced loans were probably closer to the secondary yields than to the official interest rates. By contrast, when a primary market became predominant, there tended to be a close match between official rates and primary yields. This difference can be traced to the effects of the usury prohibition and the political economy of forced loans, which implied that the city-states’ rulers had motives and means to conceal the actual weight of the debt on the popular purse.

42Consistently above par secondary prices for Genoese bonds from the seventeenth century suggest that at the time primary buyers enjoyed a mark-up. This can be partly seen as a functional requirement of a market with asymmetric information. Nevertheless this model is incomplete: Genoese bankers also invested in assets that were risky for ordinary investors, and it is unclear to what extent they played an intermediary role in the local market, where secondary prices fell in the mid-seventeenth century despite seemingly stationary risk. Political economy has been used to fill in the gaps. In this view, the actions of the Genoese capitalists were shaped by their political ties and institutional power with the consequence that their vested interest often trumped those of ordinary investors and taxpayers.

Notes

1 C. M. Cipolla, ‘Note sulla storia del saggio d’interesse. Corso, dividendi e sconto dei dividendi del Banco di S. Giorgio nel secolo XVI’, Economia internazionale, 5, 1952, p. 258.

2 C. Cuneo, Memorie sull’antico debito pubblico: mutui, compere e Banca di S. Giorgio in Genova, Genoa, Stamperia dei Sordomuti, 1842.

3 C. M. Cipolla, ‘Note sulla storia…’, op. cit., p. 260.

4 F. Braudel, The Mediterranean and the Mediterranean world in the age of Philip II, vol. II, 5th ed., Glasgow, Fontana Press, 1987, p. 700.

5 S. R. Epstein, Freedom and growth. The rise of states and markets in Europe 1300–1750, London, Routledge, 2000, p. 20.

6 D. Stasavage, States of credit: size, power and the development of European polities, Princeton, Princeton University Press, 2011, p. 123.

7 S. Homer and R. E. Sylla, A history of interest rates, 4th ed., Hoboken, Wiley, 2005.

8 M. Fratianni, ‘Government debt, reputation and creditors’ protections: the tale of San Giorgio’, Review of Finance, 10 (4), 2006, pp. 497–501.

9 J. Heers, Gênes au XVsiècle, Paris, SEVPEN, 1961, p. 170.

10 L. Pezzolo, ‘Bonds and government debt in Italian city-states 1250–1650’, in N. Goetzmann and K. G. Rouwenhorst (ed.), The origins of value: the financial innovations that created modern capital markets, Oxford, Oxford University Press, 2005, p. 155; L. Pezzolo, Republics and principalities in Italy, in P. K. O’Brien and B. Yun-Casalilla (ed.), The rise of the fiscal states. A global history 1500–1914, Cambridge, Cambridge University Press, 2012, p. 280.

11 H. Sieveking, ‘Studio sulle finanze genovesi nel Medioevo e in particolare sulla Casa di San Giorgio’, translated by O. Soardi, Atti della Società ligure di storia patria, XXXV (2), 1906, pp. 263–264; C. Desimoni, ‘Tavole dei valori’, in L. T. Belgrano (ed.), Della vita privata dei Genovesi, Rome, Multigrafica Editrice, 2nd ed. 1970 (1st ed. 1875), p. 410.

12 A. Lobero, Memorie storiche della Banca di S. Giorgio di Genova, Genoa, Tipografia Ponthenier, 1832, p. 161; H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1906, pp. 263–264; C. Desimoni, ‘Tavole dei valori’, op. cit., pp. 410–411 and 415–416.

13 D. Peri, Il negotiante, book II, Venice, Presso Gio, Giacomo Hertz, 1682, p. 96; A. Lobero, Memorie storiche…, op cit., pp. 160–161; G. Serra, La storia della antica Liguria e di Genova, book IV, Turin, Presso Giuseppe Pomba, 1834, pp. 254–255; C. Cuneo, Memorie sull’antico debito pubblico…, op. cit., p. 121 and p. 127; G. Felloni, Scritti di storia economica, vol. I, Genoa, Società ligure di storia patria, 1998, p. 453.

14 A. Lobero, Memorie storiche…, op. cit., p. 161.

15 D. Peri, Il negotiante, op. cit., p. 96.

16 C. Cuneo, Memorie sull’antico debito pubblico…, op. cit., pp. 121–127.

17 Ibid., pp. 308–311.

18 Ibid., p. 127.

19 G. Felloni, Inventario dell’ Archivio, 2013, http://www.lacasadisangiorgio.it/main.php?do=node&tag=4_4_158.

20 C. Cuneo, Memorie sull’antico debito pubblico…, op. cit., p. 121.

21 D. Peri, Il negotiante, op. cit., p. 96.

22 C. Desimoni, ‘Tavole dei valori’, op. cit., p. 411.

23 H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1906, p. 89.

24 J. D. Tracy, A financial revolution in the Habsburg Netherlands. Renten and renteniers in the county of Holland, 1515–1565, Los Angeles, University of California Press, 1985, p. 110.

25 L. Pezzolo, ‘Bonds and government debt in Italian city-states, 1250–1650’, in N. Goetzmann and K. G. Rouwenhorst (ed.), The origins of value: the financial innovations that created modern capital markets, Oxford, Oxford University Press, 2005.

26 See in particular M. Fratianni, ‘Government debt…’, op. cit.

27 L. Pezzolo, ‘Bonds and government debt…’, op. cit.

28 H. Sieveking, ‘Studio sulle finanze genovesi nel Medioevo e in particolare sulla Casa di San Giorgio’, translated by O. Soardi, Atti della Società ligure di storia patria, XXXV (1), 1905, p. 51 and p. 65.

29 H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1906, pp. 39–41, 97 and 99; J. Heers, Gênes au XVsiècle…, op. cit., p. 630.

30 G. Giacchero, Il Seicento e le compere di S. Giorgio, Genoa, Sagep éditrice, 1979, p. 257.

31 D. Peri, Il negotiante, op. cit., p. 97.

32 H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1905 and 1906; A. Molho, ‘The state and public finance: a hypothesis based on the history of late medieval Florence’, Journal of Modern History, 67, supplement, 1995, pp. 107–108; D. Stasavage, States of credit…, op. cit., pp. 119–120.

33 H. Sieveking, ‘Studio sulle finanze genovesi…’, op. cit., 1906, p. 27; S. Homer and R. E. Sylla, A history of interest rates…, op. cit.; J. H. Munro, ‘The medieval origin of the financial revolution: usury, rentes and negotiability’, International History Review, 25 (3), 2003, p. 516.

34 S. Homer and R. E. Sylla, A history of interest rates…, op. cit.; J. H. Munro, ‘The medieval origin…’, op. cit., p. 523.

35 L. Pezzolo, ‘Bonds and government debt…’, op. cit.

36 Sources: per capita weight of the debt in Florence and Venice: L. Pezzolo, ‘Government bonds and credit markets in Renaissance Italy’, in F. Piola Caselli (ed.), Government debts and financial markets in Europe, London, Pickering & Chatto, 2008, p. 19; Genoa: nominal size of the debt: G. Felloni, ‘Le attività finanziarie’, in G. Assereto and M. Doria (ed.), Storia della Liguria, Bari, Laterza, 2007, p. 133; silver content of the Genoese lira: P. Malanima, L’economia italiana. Dalla crescita medievale alla crescita contemporanea, Bologna, Il Mulino, 2002, p. 409; population: P. Malanima, Italian urban population 1300–1861 (the database), 2005, http://www.paolomalanima.it/default_files/Page646.htm.

37 J. H. Munro, ‘The medieval origin…’, op. cit., pp. 516–517.

38 L. Pezzolo, ‘Bonds and government debt…’, op. cit., and ‘Government bonds and credit markets…’, op. cit.; G. Felloni, ‘Le attività finanziarie’, op. cit.; D. Chilosi, ‘Risky institutions: political regimes and the cost of public borrowing in early modern Italy’, working paper 177/13, London School of Economics, Economic History Department, 2013.

39 F. Piola Caselli, ‘The formation of fiscal states…’, op. cit., p. 292.

40 D. Chilosi, ‘Risky institutions…’, op. cit., pp. 10–11.

41 G. Felloni, Gli investimenti finanziari genovesi in Europa tra il Seicento e la Restaurazione, Milan, Giuffrè, 1971, pp. 145, 169 and 218.

42 M. Flandreau and J. H. Flores, ‘Bonds and brands: foundations of sovereign debt markets’, Journal of Economic History, 69 (3), 2009, pp. 653–654.

43 G. Felloni, Gli investimenti…, op. cit., pp. 90–91.

44 D. Chilosi, ‘Risky institutions…’, op. cit.

45 G. Felloni, Scritti di storia economica…, op. cit., p. 176.

46 D. Chilosi, ‘Risky institutions…’, op. cit.

47 G. Felloni, Gli investimenti…, op. cit.; K. Béguin, Financer la guerre au XVIIsiècle. La dette publique et les rentiers de l’absolutisme, Seyssel, Éditions Champ Vallon, 2012.

48 G. Felloni, Scritti di storia economica, op. cit., p. 176; L. Pezzolo, ‘Bonds and government debt…’, op cit., p. 156.

49 G. Felloni, Gli investimenti…, op. cit.

Table des illustrations

Titre Figure 1. Yields on San Giorgio’s shares and on San Giorgio’s dividends with constant and variable delays, 1522–1751
Légende Notes: A = secondary market returns on San Giorgio’s shares; B = yearly rates of discount on San Giorgio’s dividends with constant delays; B adjusted = yearly discounts on San Giorgio’s dividends with variable delays.
URL http://books.openedition.org/igpde/docannexe/image/3982/img-1.png
Fichier image/png, 85k
Titre Figure 2. Official interest rates on San Giorgio’s shares (present value) and the republic of Genoa’s debt (yearly minima) and primary yields on San Giorgio’s dividends, 1625–1751
Légende Notes: B adjusted = primary yields on San Giorgio’s dividends.
URL http://books.openedition.org/igpde/docannexe/image/3982/img-2.png
Fichier image/png, 97k
Titre Figure 3. Secondary market prices of San Giorgio’s shares (yearly means, in % of par)
Légende Sources: C. M. Cipolla, ‘Note sulla storia del saggio d’interesse. Corso, dividendi e sconto dei dividendi del Banco di S. Giorgio nel secolo XVI’, Economia internazionale, 5, 1952, p. 270; C. Cuneo, Memorie sopra l’antico debito pubblico, mutui, compere e Banca di S. Giorgio in Genova, Genoa, Stamperia dei Sordomuti, 1842, pp. 309–311; J. Heers, Gênes au XVsiècle, Paris, SEVPEN, 1961, p. 630.
URL http://books.openedition.org/igpde/docannexe/image/3982/img-3.png
Fichier image/png, 36k

Auteur

Research Officer at the Economic History Department of the London School of Economics and Political Science, David Chilosi currently works on the pre-industrial economy of the wheat and the capital markets. He has recently published: ‘Risky institutions: political regimes and the cost of public borrowing in early modern Italy’, Journal of Economic History, 74 (3), 2014, pp. 887–915; (with T. Murphy, R. Studer and C. Tuncer) ‘Europe’s many integrations: geography and grain markets, 1620–1913’, Explorations in Economic History, January 2013, pp. 46–68; (with O. Volckart) ‘Money, states and empire: financial integration and institutional change in central Europe, 1400–1520’, Journal of Economic History, Sept. 2011, pp. 762–791; ‘Old wine in new bottles: civic nation-building and ethnic nationalism in schooling in Piedmont, ca. 1700–1861’, Nations and Nationalism, July 2007, pp. 417–436.

© Institut de la gestion publique et du développement économique, 2017

Conditions d’utilisation : http://www.openedition.org/6540

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search