The ‘Evolutionary’ Model of Institutional Development
We will engage head on with the question of whether premium insurance ‘superseded’ instruments like GA, but it should be noted that this question itself is, in part, founded upon teleological assumptions about the stadial evolution of commercial institutions which recent scholarship has sought to challenge: specifically, the assumption that insurance, more effective and, above all, more ‘modern’, axiomatically rendered older instruments obsolete.
1 See pp. 6–10. Business historians have increasingly sought to criticise an older body of scholarship which posited an evolutionary series of ever more efficient or rational organisational ‘technologies’, where each new development quickly superseded its predecessors. Douglass North, in particular, has been criticised for insisting on the need to chart ‘an evolutionary story of the institutionalisation of risk’, which stretches ‘from the
commenda itself … through its evolution at the hands of Italians,
to the English joint-stock company’.
2 Trivellato, ‘Renaissance Florence’, p. 243; North, Institutions, p. 127. Recent archival work by early modern historians has found that those institutional innovations considered by North and others to be inherently more efficient did not in fact lead to the rapid eclipse of older instruments.
3 Ceccarelli, ‘Risky narratives’, p. 10. In some cases, the most ‘modern’ techniques did not even enjoy widespread use until after the end of the
ancien regime, long after their initial emergence.
4 Trivellato, ‘Renaissance Florence’, p. 246. We must not risk falling into caricature here: North was well aware that the most efficient solution
in perfect conditions (i.e. free from transaction costs) did not exist in historical reality.
5 North, ‘Institutions’, p. 98. What is problematic is the suggestion that actors always availed themselves of a single, superior institution, that newer innovations displaced older ones almost completely, or more likely failed to replace them, thus dooming a society to economic stagnation. Ron Harris’s recent analysis of long-term institutional development, with particular emphasis on the importance of the emergence of the early modern joint-stock company, updates North’s evolutionary framework and renders it more sophisticated, putting emphasis on the political-economic matrices in which institutions can exist – but his schema of ‘endogenous’, ‘migratory’, and ‘embedded’ institutions nevertheless relegates institutions like GA to a timeless ‘proto’ era.
6 Harris, Going the Distance. It is perhaps telling that in an essay entitled ‘General average and all the rest’, Harris, having implicitly relegated GA to an also-ran from the outset, talks mainly about premium insurance.
7 Harris, ‘General average and all the rest’.North made another, more straightforward mistake in his analysis, namely his assumption that premium insurance in its modern actuarial form leapt fully formed from the mind of the medieval merchant.
8 North, Institutions, p. 125. This misperception has been given additional impetus thanks to several other influential historiographical trends. In 1921, Frank Knight made a now-ubiquitous distinction between immeasurable and unquantifiable ‘uncertainty’ on the one hand, and quantifiable ‘risk’ which can be priced into economic decision making on the other, though it is perhaps less often remembered that Knight’s ultimate aim in making this distinction was not a taxonomy of risk per se, but to explain profit creation in static markets.
9 Knight, Risk, Uncertainty and Profit, p. 17. In a historical context, this distinction can have the effect of suggesting that insurance was something radically ‘modern’, a watershed between one type of economic environment and another, an instrument of an altogether different nature from the risk-management strategies which preceded it.
10 Stephen LeRoy and Larry Singell Jr, ‘Knight on risk and uncertainty’, Journal of Political Economy 95 (1987), 394–406, at p. 396; Ross Emmett, ‘Reconsidering Frank Knight’s Risk, Uncertainty and Profit’, Independent Review 24 (2020), 522–41, at p. 538. In reality, before the eighteenth-century advent of the insurance company, this ostensible instrument of risk transfer was still in many ways akin to a risk-sharing instrument whose animating spirit was not widely dissimilar to that of GA. Though a few outsiders with capital to spare did dabble in insuring, the overwhelming majority of underwriters were merchants themselves: there were no specialist underwriters and very few of the merchants who insured expected to reap large profits from their underwriting activities.
11 Andrea Addobbati, ‘Italy 1500–1800: cooperation and competition’, in Adrian Leonard (ed.), Marine Insurance: Origins and Institutions, 1300–1850 (Basingstoke: Palgrave Macmillan, 2016), 47–77, at p. 62; Addobbati, Commercio, rischio, guerra, p. 126; Sabine Go, ‘Amsterdam 1585–1790: emergence, dominance, and decline’, in Adrian Leonard (ed.), Marine Insurance: Origins and Institutions, 1300–1850 (Basingstoke: Palgrave Macmillan, 2016), 106–29, at pp. 112, 120; Ceccarelli, ‘Risky narratives’, pp. 4–5; Giovanni Ceccarelli, Risky Markets: Marine Insurance in Renaissance Florence (Leiden: Brill, 2020), p. 206. The chief aim of their engagement with insurance was to hedge risks, and the same individuals would sometimes enter the market as underwriters and sometimes as buyers of insurance. While underwriters appear to have been reasonably astute judges of risk for the purposes of premiums, in the seventeenth century there were no actuarial techniques adopted for calculating risk, and a paucity of relevant information would have prevented their effective use.
12 See Lorraine Daston, Classical Probability in the Enlightenment (Princeton: Princeton University Press, 1988); Craig Turnbull, A History of British Actuarial Thought (Cham: Palgrave Macmillan, 2017); Ian Hacking, The Emergence of Probability: A Philosophical Study of Early Ideas about Probability, Induction and Statistical Inference, 2nd edn (Cambridge: Cambridge University Press, 2009). On the accuracy of early modern premium rates see Jeroen Puttevils and Marc Deloof, ‘Marketing and pricing risk in marine insurance in sixteenth-century Antwerp’, Journal of Economic History 77 (2017), 796–837. Given the reciprocal nature of the market, it has even been argued the premium is perhaps best viewed as ‘compensation’ to get round asymmetries of demand for risk coverage rather than a price, per se.
13 Addobbati, Commercio, rischio, guerra, pp. 9–10. This is not to underplay the ultimate importance of insurance, or of the innovation it represented and its eventual role in the advent of a ‘rationalistic’ modernity, but rather to guard against the introduction of anachronisms which might skew our analysis of risk management in the seventeenth century.
14 Ceccarelli, Risky Markets, p. 61.Archival evidence, moreover, is making it ever clearer that an institutional teleology is not an accurate guide to contemporary economic strategies. Early modern businesspeople used a range of institutional solutions concurrently, whether these were enforcement mechanisms to overcome agency problems or business partnership arrangements that determined their forms of commercial association.
15 Trivellato, ‘Renaissance Florence’, pp. 242–7. This chapter demonstrates that those institutions directly pertaining to sea risks, including GA, were no different. In fact, the mixed approach was even more prevalent and important in maritime risk sharing, because the various institutions employed were often complementary. Each institution offered an additional layer of repartition, and, when working in concert, a cost would be redistributed several times through several different instruments to a wide range of players, even before other risk-management tactics, such as portfolio diversification and partial ownership of assets, are taken into account. In this sense, newer institutions were superimposed upon older ones rather than replacing them. This system of risk was not solely comprised of insurance and GA: the documentation in the archive of the
Consoli del Mare amply testifies that the sea loan (a loan which involved the lender assuming not only the credit risk but sea risks as well) remained ubiquitous in seventeenth-century Livorno, though it has so far largely evaded scholarly attention outside of its medieval origins.
16 See Zanini, ‘Financing and risk’. GA’s relationship with this important institution, or rather the collection of slightly different practices which the Italians collected under the expression ‘
cambio marittimo’, will likewise be examined in detail in this chapter.
17 Raymond De Roover, ‘The cambium maritimum contract according to the Genoese notarial records of the twelfth and thirteenth centuries’, Explorations in Economic History 7 (1969), 15–33, at p. 16. GA payments could be covered by premium insurance and
cambio marittimo. Furthermore, GA played a role in its own right because it covered certain eventualities and situations which could not be covered by other risk-management institutions, a function of its non-contractual nature.
This chapter stops short of declaring that the resulting system of risk management was optimal, or even ‘efficient’. Sheilagh Ogilvie has argued that the study of institutions by economic historians has all too often fallen into the trap of making such assertions, without pausing to consider how this might be definitively proved, or even what we mean by ‘efficiency’.
18 Sheilagh Ogilvie, ‘Whatever is, is right? Economic institutions in pre-industrial Europe’, The Economic History Review 60 (2007), 649–84, at pp. 656–8. What this documentary evidence can demonstrate is that actors, contrary to North’s evolutionary model, did in fact avail themselves of a variety of risk-management institutions simultaneously, and that by using a variety of techniques which were complementary they achieved a far greater spread of risks than would otherwise have been possible. It also demonstrates that GA and sea loan were possessed of unique advantages and capabilities which responded to the context of seventeenth-century Mediterranean trade. This in turn suggests that their continued use was not the result of historical contingency, path dependency, or even high transaction costs, but was at least a sensible response to the conditions in which actors found themselves.