Andrew Foster (Fordham University)
Embedded within the 4th century Athenian maritime contract preserved in Demosthenes’ Against Lacritus is evidence that maritime lenders charged different rates of interest at different times and destinations based upon the relative risk of each prospective journey:
Androkles the Sphettian and Nausikrates the Karutian lent to Artemos and Apollodoros the Phasalitians 3000 silver drachmas [to sail] from Athens to either Mende or Skione and from there to the Borsporos, and, if they wish, as far as Borysthenos provided they sail on the west coast and back to Athens, at the rate of 225 drachmas per 1000 drachmas borrowed [22.5%], but if they sail after the rising of Arktouros from Pontos for Hieros, the interest will be 300 drachmas [30%] (Dem. 35.10-1)
Scholars have recognized that lenders charged higher interest rates to a borrower as a “risk premium” and that various rates for different itineraries executed at different times formed a sophisticated form of relative risk management (Calhoun 1930: 575-76). There is also widespread consensus that, as Millet (1991: 192) has observed,
. . . it is presumed that for maritime lenders to last long in the business they would need to price risk accurately: Given the high risk of maritime credit for the lender, they would need to have the knowledge to assess the relative dangers of a voyage, and arrive at an appropriate rate of interest.”
In this paper I will argue that lenders did indeed variably price money in order to manage the risks associated with maritime trade, but that lenders lacked the ability to price risk accurately enough to ensure that variable interest rates alone effectively managed risk. Rather, it was essential for maritime lenders to identify and affiliate with borrowers and agents who shared similar risk preferences and voluntarily pooled information relevant to risks and returns of a particular venture. Without taking measures to homogenize the risk preferences between lender and borrower or prevent information asymmetries from prevailing, lenders would regularly ration credit and/or promote adverse selection and thereby misallocate credit (Hoff and Stiglitz 1990: 237-8)
In both Dem. 35 and [Dem.] 56 the lender accuses the borrower of arbitraging their loans (Dem. 35.36; [Dem.] 56.29). In both these cases the borrowers arbitraged their loans to finance less risky and potentially more lucrative, clandestine side projects. These traders’ secondary lending -- whether actually done or only plausibly asserted -- illustrates how lenders are thoroughly dependent upon borrowers to share information in order to allocate credit efficiently.
Additionally, if a lender fails to assess accurately a potential borrower’s risk preferences, a lender would inevitably underwrite a risk insensitive borrower (adverse selection) or price too high for a relatively risk averse borrower (credit rationing). In sum, just as traders managed risk by plying familiar routes and trading in a limited range of goods repeatedly (Morely 2007: 33) lenders would better manage risks by lending to such traders who also shared information and risk preferences.