The Impact of GA on Underwriters

A sense of GA’s relative importance and weight for an insurer can, however, be gleaned from merchant accounts books dealing with insurance, even if these have rarely survived and cannot generally provide us with serial figures over long periods.2 Andrea Addobbati, ‘War, risks, and speculation: the accounts of a small Livorno insurer (1743–1748)’, in Phillip Hellwege and Guido Rossi (eds), Maritime Risk Management: Essays on the History of Marine Insurance, General Average and Sea Loan (Berlin: Duncker & Humblot, 2021), 161–88, at p. 165; Lewis Wade, Privilege, Economy and State in Old Regime France: Marine Insurance, War and the Atlantic Empire under Louis XIV (Woodbridge: The Boydell Press, 2023), p. 131. The data presented in this analysis pertains to the final third of the century, and thus the part of our period in which the use of insurance was most extensive. The sources in question are two account books preserved in the Quaratesi collection in the Florentine state archive. These are the records left by this ancient Florentine family with a long history of commercial activity since the Middle Ages, with some family members living in both Pisa and Livorno.3 Giulia Camerani Marri, ‘Archivio Quaratesi’ in ‘Notizie degli archivi Toscani’, Archivio Storico Italiano 118 (1960), p. 368. The family were in fact responsible for one of the more famous artistic representations of a jettison at sea, ‘St Nicholas saving a ship from the tempest’, which formed part of the ‘Quaratesi Polyptych’. This altarpiece was painted by Gentile da Fabriano, and was originally housed in the church of San Niccolò Oltrarno.4 See Lionel Cust and Herbert Horne, ‘Notes on pictures in the royal collections. Article VII – The Quaratesi Altarpiece by Gentile da Fabriano’, The Burlington Magazine 6 (1905), 470–5. See also Richard Goldthwaite, ‘The practice and culture of accounting in renaissance Florence’, Enterprise & Society 16 (2015), 611–47, at p. 631.

The Quaratesi archive ranges from the fourteenth to the nineteenth century and is mostly comprised of documents related to the family’s mercantile activities and the administration of their patrimony.5 Marri, ‘Archivio Quaratesi’, p. 368. At least two of the seventeenth century account books contain accounts related to insurance. The first is a ledger maintained by one of the Quaratesi themselves though it is unclear who exactly: the author at one point makes reference to ‘Girolamo Quaratesi, my brother’ but his own name is never mentioned.6 ASF, Quaratesi, 357. The account was made in Livorno, detailing underwriting carried out by this Quaratesi on his own account, and that carried out on behalf of his associate, Baccio del Beccuto, who likewise hailed from a prominent Florentine family active in trade across the early modern period.7 Ricardo Court, ‘“Januensis ergo mercator”: trust and enforcement in the business correspondence of the Brignole family’, The Sixteenth Century Journal 35 (2004), 987–1003, at pp. 992–4. This first book covers the period 1670–85 with accounts for Beccuto for 1670–9. Beccuto appears to have been a passive partner in this first period, and the insurance policies signed in his name replicate exactly those underwritten by Quaratesi, though often for smaller amounts (usually half of that underwritten by the managing partner). The second book was maintained by Beccuto alone, began in 1691 and continued until 1695 when Beccuto ‘passed on to a better life’ in the early hours of a February morning.8 ASF, Quaratesi, 261.

The first section of each account records insurances made. Each entry details the buyer of the insurance, the nature of the insured item, the port of origin, the port of destination, and the name of the relevant ship and its master. The value underwritten is recorded to the left of each entry, and the premium to the right: all premiums are slightly less than a round number, representing the fact that the amount received was reduced by the fee paid to the insurance broker. Journeys that finished successfully are marked with an ‘A’; those where there was a pay-out are marked instead with a ‘P’. In the second book, a total is made at the end of each year, detailing profit and loss: the premiums are totalled and set against the complete losses and the Averages (sinistri e avarie) which Beccuto paid out for, as well as another fee for his broker. The same calculation is performed in Quaratesi’s book but at irregular and apparently arbitrary intervals. Quaratesi also transferred small sums out of his insurance account into other accounts on a regular basis, making his figures less straightforward to analyse.

The second part of Beccuto’s book also provides a separate breakdown of these losses and Averages, as well as the storni – refunds of the premium when the voyage did not go ahead, or when it was discovered that the vessel or cargo had been over-insured. This breakdown allows us to analyse the relative weight of Averages as a proportion of overall outgoings. In line with the slightly more chaotic nature of Quaratesi’s account, a similar breakdown is not always included in his book, meaning that only the years 1671–8 can be fully analysed. Quaratesi also employed a somewhat irregular accounting technique: rather than balancing his ingoings (premiums) and outgoings (losses, Averages, and storni) on a yearly basis, he simply interrupts his list of policies at random intervals to record the ‘losses and Averages paid up to this date’.9 E.g. ASF, Quaratesi, 357, f. 16r. So we are unable to analyse his account on a year-by-year basis, and instead are forced to divide the account up into three uneven periods: 23 June 1671 to 18 July 1672; 26 August 1672 to 7 September 1674; and 20 September 1674 to 31 October 1678. Quaratesi, unlike Beccuto, did not note the amounts he underwrote when he accepted a premium.

In total, we can therefore analyse the frequency and relative weight of Average pay-outs for the years 1671–8 and 1691–5. It should be remembered at this point that there is no distinction made between GA and PA in the source, which considers only losses from ‘Average’ as a whole. Table 4 details a breakdown of the pair’s outgoings in the period 1671–8 (Quaratesi recorded the losses for both partners as a single figure). Table 5 captures the whole of Beccuto’s underwriting activities, showing the total premiums accepted and the losses broken down by type, including complete losses (perhaps sometimes mitigated by salvage), returned premiums, Averages (with no distinction made between PA and GA), and other small, miscellaneous outgoings such as payments to agents or attorneys.

We can see that the underwriting directed by Quaratesi in the 1670s was not as extensive as that carried out by Beccuto in the later period. The overall weight of Average on the associates was modest, just under 15 per cent of total losses. Yet the percentage of losses deriving from Average in the 1670s account varies quite considerably: 1.31 per cent in the first period of just over a year, 38.19 per cent in the second period of around two years, and 3.94 per cent in the third period of around four years. There is no way for us to establish what the typical relative weighting of GA and PA might have been: while we can be fairly certain that most of the GAs left traces in the Consular archives, there are very few cases involving PA, and it seems likely that the majority were dealt with privately.10 E.g. ASP, CM, AC, 319-6 (5 February 1669); see Dyble, ‘Divide and rule’, pp. 371–2. We can nevertheless be sure that the impact of GAs was even less pronounced than this figure suggests. The period in question was perhaps not the most turbulent faced by Livornese insurers, but nor can we say that this was a period of relative calm, in which Average contributions might have been lower than usual. The Franco-Dutch War (1672–8) was ongoing for most of the period, a conflict which included the Third Anglo-Dutch War (1672–4), and both of these would have had an effect on Livornese underwriters, raising premiums but also increasing the rate of GAs and PAs.11 See Wade, Privilege, Economy and State, p. 132. It is thus unlikely that the figure of 15 per cent vastly understates the weight of Average losses generally sustained by underwriters.

The underwriting carried out by Beccuto in the 1690s was of a different level of magnitude. Quaratesi received a mean average of just under 130 pieces a year in premiums, with Beccuto generally insuring half that amount on each risk. From 12 June 1670 to 31 October 1678, Quaratesi accepted a total of 1,085 pieces in premiums. Beccuto accepted more in premiums each year in the 1690s than Quaratesi had in the entire period covered by his account book. We can also be fairly sure that Beccuto was re-entering the insurance market after a period of absence. The first year of activity recorded by the book details some refunded premiums, but no total losses or Averages. This reflects the delay that occurred between the underwriting of policies and the eventual insurance claims made upon them. Had this account book been a continuation of an existing account, we would expect to see losses and Averages from previous years carried over. Beccuto, like many investors, could have been persuaded to reengage more extensively in insurance by the intensification of the Nine Years War (1688–97), which, like most conflicts, increased premiums and tempted new and part-time investors into the market with the promise of large windfalls.12 Addobbati, Commercio, rischio, guerra, pp. 92, 116; Addobbati, ‘War, risks, and speculation’; Ceccarelli, Risky Markets, pp. 103–20. The war had a considerable impact on maritime trade, however, as the English and Dutch abandoned a traditional indulgence to neutral commerce and agreed to attack all vessels of whatever flag sailing to French ports or carrying the goods of French subjects.13 George Clark, ‘The Nine Years War, 1688–1697’, in J.S. Bromley (ed.), The New Cambridge Modern History. Volume 6: The Rise of Great Britain and Russia, 1688–1715/25 (Cambridge: Cambridge University Press, 1970), 223–53, at pp. 234–5. English merchants were later to demand greater protection from the Royal Navy in 1690 on account of the number of ships they had lost in the Mediterranean.14 Clark, ‘The Nine Years War’, pp. 238, 244.

After an initial profit resulting from the aforementioned time lag, Beccuto consistently made a loss each year. Overall, he lost more than 800 pieces over this three-and-a-half-year period. This could conceivably be a result of extending his activities, as he took on more and more risks which would have been eschewed in the earlier, more cautious phase. Alternatively, he could simply have been unlucky. Either way, the Averages represent a slightly diminished though not altogether dissimilar proportion of losses: 9.23 per cent.

When we examine Beccuto’s outgoings in more detail, however, another interesting qualification emerges. Beccuto was very rarely insuring ships, freight, or sea loans – the three ways in which he could have been covering risks run by the shipmaster and/or ship-owners. Of the 104 pay-outs recorded in his accounts, only three were the result of insurance made on a ship, and only two of insurance made on sea loans. Of these five pay-outs, none related to an Average. As Ralph Davis remarks in relation to seventeenth century England, and as Lewis Wade has found more recently in the case of seventeenth-century France, merchandise was insured far more often than ships were.15 Ralph Davis, The Rise of the English Shipping Industry in the Seventeenth and Eighteenth Centuries, 2nd edn (Newton Abbot: David & Charles Publishers, 1972), p. 87; Wade, Privilege, Economy and State, p. 133.

GA was thus providing protection to ship-owners from risks which insurance was not yet covering: the insurance ledgers do not tell the whole story. Ship-owners were of course already hedged against such risks to a certain extent through the practice of shared ownership. Investors generally owned only a part of a ship in order to create a safer portfolio, but it should be noted that an ownership share of less than an eighth was extremely uncommon, at least in an Italian context.16 Davis, English Shipping Industry, p. 83; Piccinno, ‘Genoa, 1340–1620’, p. 40. GA thus provided a second line of defence. Once again, we find GA of particular importance for the transport side of the commercial sector. This risk-management dimension can be set alongside other bonuses for the transport sector, such as the opportunity to repartition routine costs, which we have already encountered when examining consolati and in the case of the Alice and Francis.