1. A contract for private trade finance, aside from the bill of exchange

As the contract mobilized capital to procure silver to further trade elsewhere, with a turnover in the order of several months or years, the contract resembles a venture capital. Yet, when looking into the currency component of the contract, e.g., the specification of the specie on which to be reimbursed, respondentia also looks more like a forward swap; it dealt with mark-ups, like spot and future prices of goods whose quality was difficult to price—like sugar and diamonds.Footnote ² Irrespectively, the contract clearly was suitable to finance trade among economies performing with different cash means of payment and to obtain silver specie—i.e. money—in Mexico, in Cadiz or Manila—as shown by Lamikiz’ and Ruiz Stovel and Ruiz Moreno’ articles in the collection. In these regards, it was clearly an “importer” type of finance, that mitigated costs derived from exchange and currency risk and realized profit from arbitrage when dealing foreign specie within private money markets—as discussed below.

Looking into notarized contracts in the thirteenth-century Genoa and the twelfth- and thirteenth-centuries trade with pepper in Byzantium, De Roover (Reference De Roover, Postan, Rich and Miller1963, p. 54) identified three types of sea loans which superseded the long-established equity partnership commenda as maritime commerce expanded: one was the “ordinary” sea loan (foenus nauticum). This loan was “unsecured save for a general lien on the debtor’s property.” Another “called in medieval England the respondentia” was a loan in which the “principal and increment must be paid even though the ship perishes provided the cargo to be safe”; and a third type was the bottomry loan, which were usually secured by part or the whole of the ships or its freights subsequently. He also highlighted another contract that increasingly replaced the ordinary sea loan and was becoming popular by the thirteenth century: the cambium maritimum. In a later piece, De Roover (Reference De Roover1969) could not neatly establish whether this was an exchange or a credit transaction. If the latter, he argued as “transfer instrument it could serve as an international long-distance settlement” (p. 20); if the former, an operation to disguise a loan, it was then a credit instrument for trade”—although both were probably not mutually exclusive. More importantly for the argument on this essay, as de Roover pointed out, “It anticipates future developments in the money market” (p. 22).

Respondentia differed from any other contemporary commercial and financial instruments in one critical aspect, which economic historians often disregard. Insofar as there was a change (permutatio) of money (pecunia) it often hid the interest and true nature of the contract as a loan, so historians alleged that it was favored to avoid the usury laws. However, the covenant enabled long-distance exchanges of goods for money, by specifically establishing the species with which the capital was to be paid back. Therefore, respondentia, as a version of cambium maritimum, always involved an operation of exchange (Lo Basso, Reference Lo Basso2016b).Footnote ³ This made respondentia also very distinct from bills of exchange, which sought to minimize the use of bullion in settling international balances. Creditors’ persistent demand for specie (silver cash) in Asia and Europe, resulting from their intermediation in the world’s trade, affected the price and supply of bills. This made the band for the specie point consistently very wide indeed—i.e., little room for arbitrage with bills—so silver continued flowing East from America to Europe and from Europe to Asia, and bills took longer to serve as remittance instead (Flandreau, Reference Flandreau2004; Nogues, Reference Nogues and Battilossi2020; Irigoin et al., Reference Irigoin, Kobayashi and Chilosiforthcoming). Arguably, this made respondentia purposefully designed to carry an exchange transaction in bullion markets that were not (yet) integrated.

As an advance of money or goods, respondentia was a free private arrangement for trade finance among disparate mercantile agents in remote locations that dealt with foreign monies better than otherwise. A bit hesitantly, Denzel describes the préstamo marítimo as “not a classic bill of exchange transaction” but rather an «emprunt contracté», with a questionable negotiability … (an instrument) that served (alone?) to acquire money” (2010, xlvii; his emphasis). Thus, it was also a collateralized loan, though not a direct financing vehicle, which, in its form of debt for the full value of the contracts, served to acquire foreign exchange. It resembles a modern derivative, the foreign exchange or currency swaps—i.e., two parties exchange currencies at the current spot price and commit to reverse the exchange at some pre-agreed future date and price (the forward rate) in the established foreign currency (usually the USD dollar). It involves an implicit interest payment in the harder of the two currencies—which can be persistent and conform “puzzlingly” large extra expenses, justified by the “dominant currency status” of the foreign money as the preferred international currency of the time (Gopinath and Itskhoki, Reference Gopinath, Itskhoki, Gopinath, Helpman and Rogoff2022).Footnote ⁴ The silver coined in Spanish America enjoyed the status of dominant currency until the 1820s.

Unlike bills, respondentia had no usance (or not always) but a fixed customary period following the arrival at the port for the settlement, beyond which interest rates may be charged.Footnote ⁵ In principle, there was no balance left; lending was settled with the payment as the taker was bound to do it in full—and in cashFootnote ⁶—after the completion of the journey. There was no room for other transactions as re-change, although some contracts in a round trip over the Atlantic might have involved two distinct bonds with different types of money or rates for each leg.Footnote ⁷ Neither respondentia bonds were negotiable, or discounted, nor were they transferable or endorsed. Therefore, they made no real contribution to enhance liquidity in the market. There was no protest, however, if the borrower refused to pay the bond was not sent back to the issuer; it opened the path to execute the guarantee, or the guarantor agreed. It was a two-party-direct contract, so there were no additional costs—i.e., commissions paid for intermediation.Footnote ⁸ Like bills, they were also a foreign exchange transaction that indeed involved different monies, while respondentia did not conceive a conversion of currencies.

Bills involved two (or more) different currencies and monies of account as exchanges meant a deal with one another by each buying a bill in the local market and paying a sum denominated in a local money of account for the delivery in another market of an equivalent amount in the respective money of account. They were a draft, order or direction of one person addressed to another person requesting to pay a sum of money to a third person. In the case of foreign bills drawn in the currency of one country upon a person residing in another—hence performing with a different currency—yet to pay the third one who might (or might not) be in the same currency regime.Footnote ⁹ The negotiability of the bill involved different currencies and means of payment. That is a conversion of currencies to monies of account and vice versa. The bill was also for the absolute—unconditional—payment of a sum of money and not on goods or stocks “not to depend on any uncertainty or contingency.” It was too for the payment of money, but “it matters not that the denomination of money specified has, because the value of each kind can be ascertained.”Footnote ¹⁰ Respondentia instead, was a covenant between two parties that allowed the return in cash (specie) to be certain for the lender ab initio—less often it established the exchange rate at which the local currency was to be priced in the return; in either case, the contract also reduced the so-called exchange risk.

2. Terms, conditions, and rates

The idea of very high risks in long-distance trade has persuaded historians that the high “interest rates” charged on the principal were directly related to such hazards. Hence, the characterization of respondentia as a «high-profit/high-risk» investment. Rates of over 50 or even 70% were frequent in the seventeenth and early eighteenth century’ Peru trade as recorded by Quiroz (Reference Quiroz1993). However, the commercial risk distinctly fell on the borrower. As the rate—indeed, the premium or reward—was established ex ante with the initial capital outlay, the borrower must have been confident that the ventures would yield even higher returns, or enjoyed much better information, as Costa argues in her article in the collection. Arguably, the premium might also be considered the upper bound rate at which borrowers were willing to pay for working capital rather than an indication of risk or the opportunity cost of capital. Everywhere, premiums quoted were systematically much higher than the customary local interest rates (Flandreau et al., Reference Flandreau, Galimard, Jobst, Nogués-Marco, NEAL and Atack2009, appendix A, table 6A.1).Footnote ¹¹

Historians tend to correlate rates with distance, so some see a time dimension of the cost of capital in it. Premiums, however, were very similar in the Cadiz-Veracruz and in the Acapulco-Manila voyages, despite different business organization and very different lengths of time at sea, as shown in the introduction to this special issue. In the 1660s, Genoa’ rates for Lisbon, Smyrna or Athens were 80% higher or double than those for Alicante or “coast of Spain,” despite comparable distances. Moreover, these rates pale before premiums for “New Spain” via Cadiz (Lo Basso, Reference Lo Basso, Giuffrida, Rossi and Sabatini2016a, pp. 155–58). Europeans (non-Spanish) charged 15–16% from Flanders and Amsterdam to Cadiz in the later seventeenth century, but rates jumped to 40 and 60% if capital continued to “the Indies” or the River Plate (Everaert, Reference Everaert1971, p. 5; Freeman, Reference Freeman2020). The difference persisted in the 1760s in loans from Nimes, which paid 18–20% for Cadiz and 30% for “America” (Chamboredon, Reference Chamboredon2016). In London, Blackwell lending charged fixed monthly rates for premium “and in proportion” for a maximum of 36 calendar months; but rates varied between five and 48 pounds a month for bonds of similar amount and terms for the delays to clear after arrival (Martin’s Bank 009–0057). Seemingly, premiums did not have a clear time component, and they appear to relate to expected returns at destination.

Additional—with clearly separate and timed structured rates—interest was charged for delays when the clearing went beyond the established time post arrival. Whereas this never applied in the liquid Manila market, surcharge rates appeared later in Cadiz—at a time when returns on the loans were falling. Clearing took longer in bonds to Mexico, once the fleet and the termed fairs ceased by the mid-eighteenth century. Individual arrivals carried smaller cargoes so had lower values. They were more irregular but also much more frequent, raising the risk of gluts in the rich but shallow colonial markets for demographic and geographical reasons; selling the goods required a longer time to settle. Thus, sales were often consigned to the next season to hedge against low spot prices and losses from overstocking as flows of similar goods from Acapulco and Cadiz kept pouring over. Postponements of the clearing added interests to the principal and premiums. Elsewhere, where additional interests were charged on such delays, higher premium rates bore no relation to the (approximate) current local rates for alternative investments. In Asia, these moved around 0.75% per month in Surat and Bombay (Chaudhury, Reference Chaudhury2015, p. 91; Haider, Reference Haider1996), 1.5% in Canton (Van Dyke, Reference Van Dyke2011: 45), or an annual 7–10% as in Macao (Dos Guimaraes Sa, Reference Dos Guimaraes Sa2003; Souza, Reference Souza2004, 186).Footnote ¹² In Europe, rates were at a monthly 0.5–1% in Cadiz, 1.25% in Genoa, and 12% per annum in Flanders. Everywhere, much higher premiums seem independent from the local cost of money, so they cannot be an indication of its stock available.

Historians point to rates overshooting during wars as the foremost risk to trade; however, the effects are ambiguous. In some cases, there were provisions for increasing premiums if there was a probable threat—as in the original subscription of the French Compagnie des Indes stock (Recueil …). A specific clause about «risk of (government) expropriation» appeared in 1640s Dutch bonds for (illegal) journeys to the River Plate; they seem highly unusual, and, in this case, the risk was passed explicitly on to the borrower (Freeman, Reference Freeman2020). It is unclear whether the risk seems more associated with the impact of war in Europe than at sea or overseas. This affected risk—and premiums—differently. Reporting to his clients François Tronchin, a broker based in Geneva, made very clear the direction of trade: prices—and naturally the expected rates of return—were heavily dependent on the relative abundance of silver at destination.Footnote ¹³ Thereby, the risk calculation was always to avoid glutting markets in America (Baskes, Reference Baskes2013). War effects are not visible in the case of Manila, shown above in Figure 1. Losing the Galleon Covadonga to British Admiral Anson near Manila in 1743—laden with 1.3-million-peso coins (c. 32 tonnes) and about 35 tonnes more in bars (Cushner, Reference Cushner1971)—ought to be a blow to lenders as correspondencia returns never arrived, creating a credit crunch.Footnote ¹⁴ Yet, rates did not change and jumped only once silver laden galleons returned. Therefore, premiums do not exactly capture risk from war incidents; if any, premiums seem directly related to the availability of silver.

Figure 1.

Premiums in Atlantic, Pacific and Indian Oceans and China seas (1715–1823), in percentage.

Sources: For Manila: Ruiz Moreno (2023, p. 227); for Brazil and Asia: Costa Freire and Pinto De Albuquerque (2022); for Copenhagen: Glamann (1949, table 6).

There were means to mitigate the exposure to risk, however. Lenders apparently spread the bonds in many ships—even if sailing in a fleet—and in many small amounts. It was the same among lending to junk merchants in Canton (Van Dyke, Reference Van Dyke2011, p. 153), in the Genoese lending in the Mediterranean, as in the Spanish and French deals in the Atlantic (Carrasco-Gonzalez, Reference Carrasco Gonzalez1996 Appendix A-1; Chamboredon, 1995, pp. 704–707). Clearly, respondentia did not favor economies of scale. Similarly, the contract was non-negotiable nor transferable, so the contribution to liquidity was limited to «acquire money» as Denzel suspected. Without enhancing capital pooling and without increasing returns to scale, the contract does not seem effectively designed to favor productivity improvements.

Premiums neither did seem to bear a particular relation to the nationality of borrowers or the organization of the business. Frequently rates even varied for different transactions on board of the same ship (Carrasco Gonzalez, Reference Carrasco Gonzalez1996, p. 94, Lamikiz, Reference Lamikiz2023, p. 5). It is difficult to see a time component in the premiums, as interest were charged separately, so it cannot price the opportunity costs of capital. The round sail from Cadiz to Veracruz in Mexico took roughly six to eight months, and from Cadiz to Peru, the Acapulco galleon took over a year-long, non-stop round trip set by the Monsoon. Yet premium rates for main Mexican ports on the Atlantic and the Pacific—were very similar throughout the eighteenth century (Bernal, Reference Bernal1992; Ruiz Moreno, Reference Ruiz Moreno2023). Interestingly, both ports were also main outlets for Mexican silver. Lending at respondentia charged a premium ex ante defined as a fixed rate of return, which, being an advance of capital, suggests it was directly related to the expected profits from that trade.

Reportedly, premiums declined over the eighteenth century; the falling (non-coeval) trend over the century repeats in all cases of European commerce across the Atlantic by 1800, other than in trade with Asia through the Indian and Pacific Oceans. It is noteworthy to remember that China’s import of silver continued steadily until the mid-1820s (Irigoin, Reference Irigoin and Battilossi2020). Figure 1 charts a rare longitudinal data on respondentia rates in different long-distance routes of the time, e.g., the Manila lending for Acapulco, the Portuguese in the Atlantic and India trades, and the Danish trading in Canton—to outline the trend (data is too scattered to satisfactorily compare levels) that premiums followed in the respective routes.Footnote ¹⁵ As said it is not feasible to reconstruct any robust series of premium rates; however, the results are consistent with qualitative impressions from the literature.

Throughout the eighteenth century, rates remained high in the Indian and Pacific oceans while declining steadily over the Atlantic, clearly the most competitive of these trades (De Vries, Reference De Vries, Flynn, Giraldez and Von Glahn2003; Quiroz, Reference Quiroz1993, p. 100). The case of Lisbon confirms that they did not relate to the place where capital originated; rates for both “Indies” started at about the same level, but premiums diverged with the destination over time. Arguably, higher and sticky premiums for all contracts in Asia might reflect the persistent premium on silver cash there and the room for arbitrage gains in Asia.Footnote ¹⁶ Lastly, similar agents contracted capital at different rates depending on the returns of trade involved, somehow independently of the capital endowments available, the business model, the length and hazards of the journeys and even the “national/ethnic” origin of the parties.Footnote ¹⁷

3. Global use and foreign currencies

The Japanese word for a similar contract—nagename—means (lit.) “trust in silver” because it was the specie involved in the exchange for silk in China in the seventeenth century. In November 1641, two Dutch VOC agents invested 300 (Spanish) “reals of 48 stuivers each” on a ship owned by a Chinese going from Batavia to Palembang—Sumatra. The loan was termed with the journey duration estimated in 3 months; the premium agreed was 30% plus 2% of interest on the total amount (principal and premium)—i.e., 390 reais—for each month in arrears after the arrival.Footnote ¹⁸ Therefore, the prevalence of respondentia in the Spanish overseas trade should not be a surprise. There was no larger «fountain of silver» for the world at the time than Spanish America’s—replacing Japan as a source of China’s silver imports after 1680, and even behind Dutch ventures in Asia. Spaniards used the same instrument in both the Atlantic and the Pacific routes to procure silver in Spanish America. Whether it was on the—allegedly safer—armed fleets on the Atlantic up to the 1740s, on individual ships thereafter, or on the single large galleon that made the Pacific line, cambio maríitimo or correspondencia (as the contract was known respectively) was their mode of trade finance (See appendix). Originally, capital advanced in European or Asian merchandize flowed to America, although by the later seventeenth century, domestic capital advanced in bullion and specie flowed from America to Europe, bypassing trade fairs at Portobello (Suarez, Reference Suarez1995; Quiroz, Reference Quiroz1993, pp. 101–111). A century later colonial merchants invested in respondentia themselves in Cadiz for Lima (Lamikiz, Reference Lamikiz2023) and for Mexico (Del Vallen Pavon, Reference Del Vallen Pavon2022). In all cases, deals were notarized to establish claims and cessions. Yet, respondentia was a fully private deal over the Atlantic and the remit of institutional lenders over the Pacific. Spaniards employed other instruments in their Atlantic trade (one-off companies—limited partnerships, book credits, consignments, direct sale by representative agents with powers of attorney, etc.) with foreigners or nationals, but always as second to respondentia which was the «only game in town» in the Pacific run (Ruiz Moreno, Reference Ruiz Moreno2023; Mesquida, Reference Mesquida2018). None of those instruments intended to substitute cash remittance, as indeed cash, i.e. coined silver, was the main Spanish American export at least until the 1790s, and the most valuable Asian import. Yet there were no fiduciary means of payment or exchange in circulation in Spanish America until the very late eighteenth century (Perez Herrero, Reference Perez Herrero1988; Lana Berasain, Reference Lana Berasain2016); and business with “riesgo de tierra” (overland bonds) were rare inside Spanish America.Footnote ¹⁹

Respondentia—in their own national variation and name—was also the instrument that other Europeans used in the re-export trade through Cadiz. Dutch private merchants used it in the River Plate trade in the 1660s (Freeman, Reference Freeman2020), Flemish and Genoese did it in Mexico and Peru later (Everaert, Reference Everaert1971) and so did the Portuguese in Brazil (Costa Freire, Reference Costa Freire2003). As Cadiz’s main trading partners, French houses brokered goods and capital into the Atlantic commerce on respondentia (grosse) throughout the eighteenth century. They used it in their own direct (and legal) trade to South American Pacific coast at the turn of the eighteenth century (1698–1725) (Lespagnol, Reference Lespagnol1992/1997, Malamud, Reference Malamud1986), and with India throughout. Prêt á la grosse—as the instrument was known in French—served to finance the re-export of linen and silk textiles to the Caribbean and South America (Chamboredon, Reference Chamboredon1995, Reference Chamboredon2016) where commerce procured piastres (silver coins) with which in turn to procure cotton and silk piece goods from India (Dermigny, Reference Dermigny1960). However, these houses used bills in returns for sales and investments within France.Footnote ²⁰ This dual use of respondentia outside Europe and bills within was equally a practice of Catholic (e.g., the Magons of St. Malo) and Protestant (e.g., the Forniers of Nimes) merchants alike—which challenges the assumption of the contract’s special purpose of avoiding usury laws. More importantly, this stark distinction in the choice of instruments used in one and another trade by the same agents suggests a differentiated financial strategy.

The instrument was not confined to New World’ trade either. It was also abundantly present in European commerce in Asia, especially. Scholars of various trades repeatedly register its use, from the earlier case of Portuguese trade in Nagasaki until the 1630s, to business in Goa, Macao and India in the eighteenth century (Oka, Reference Oka2001, dos Guimaraes Sa, Reference Dos Guimaraes Sa2003; Souza, Reference Souza2004. European “country traders” and “peddlers” in Burma, India and China (Dijk, Reference Dijk2001; Tomlinson, Reference Tomlinson2002; Van Leur, Reference Van Leur1967; Steensgaard, Reference Steensgaard1974) too issued respondentia in their own businesses since the mid-seventeenth century, as Amoy’s merchants did in Siam (Viraphol, Reference Viraphol1977, pp. 126–27). Armenians and Malabari from Madras, at the service of the EIC (Arasaratnam, Reference Arasaratnam1979; Watson, Reference Watson1978) or in partnership with Dutch or French merchants, all they traded in Manila under respondentia in the eighteenth century (Quiason, Reference Quiason1966). French merchants from 1702, and Danish and Swedish from the 1730s, issued respondentia in their Canton trade (Morse, Reference Morse1926–29, vol. II, pp. 115, 149; Van Dyke, Reference Van Dyke2005). Even Boston’ houses with business in India in the 1800s (Porter, Reference Porter1937/Reference Porter1969, vol. I, pp. 53, 66, 593, 599, 602, 644, 672, 680, 686; vol. II, pp. 859, 907, 929, 1137) and British merchants did as late as the 1830s Shanghai. (GBR/0012/MS JM/F5/2,5,10,12).

Allegedly, most of the silver obtained through the large intra-Asia trade also eventually terminated in China (Cheong, Reference Cheong1970). The instrument was particularly suited to procure Spanish dollars at Manila, Mokha or Jeddah (Nadri, Reference Nadri2008:67–69) for further deals in India and Canton (Reference MorseMorse, 1926–29, vol. V, pp. 73, 111, 149). Private merchants in India advanced cash against respondentia bonds “redeemable upon maturity only in cash and only in Canton” (Cheong, Reference Cheong1971, p. 90; IOR/L/AG.1). They preferred it as means of remittance over Company bills of exchange on London when the exchange rate was not convenient (Furber, Reference Furber1948) so private bills did not fully replace respondentia (IOR/H/795 fs 14, 18, 32; Bowen, Reference Bowen2010, p. 465) until the 1830s.Footnote ²¹

A few relatively large institutional lenders of respondentia operated in Asia: the Jesuits, who early brokered silver from Japanese rulers’ trade in China for silk, and the EIC itself. The latter lent and borrowed under this contract—between September 1737 and July 1748Footnote ²² the company lent about a million-pound sterling at rates between 18% and 35% to her captains (IOR/H/449 fs 84,84 vta,101–03,245). In Portuguese Macao, an arrangement between the city Senate and the local Casa de Misericordia confraternity so funded private commerce in the Indian Ocean and China through the eighteenth century (Guedes Cosme, Reference Guedes Cosme2020; Subrahmanyan, Reference Subrahmanyan1995). A similar scheme was developed in Manila since the 1680s where legacy funds amassed large sums of coined silver and became, probably, the most specialized lender and largest provider of working capital for trade within Asia. There, the local Misericordia (House of Mercy) and religious orders, to a lesser extent, acted as investment funds channeling finance to the Acapulco commerce and subsidiary trades from China and Coromandel for re-exports to Mexico (Mesquida, Reference Mesquida2018; Yuste Lopez, Reference Yuste Lopez, Yuste Lopez and Pinzon Rios2019). Chinese traders, Armenians from Madras and even the EIC itself, tapped into such funds (IOR/H/606 1785–1798). Everywhere, the legal features of the contract binding lenders and borrowers into a limited partnership of sorts barely differed (see the Appendix). Likewise, (other than for Acapulco) the sums involved were relatively small, which lured all sorts of borrowers. This cross-cultural trade feeding into the intra-Asian commerce had different risks, transaction costs, and mostly different returns, so premiums varied among Asian destinations (Davies, Reference Davies2012, pp. 214–15; Prakash, Reference Prakash2007). Across trade in Asia, rates systematically appear to be higher when lending was for China. Roughly, in 1750s and 1760s’ Macao, the rate for China was 20–25% (Dos Guimaraes Sa, Reference Dos Guimaraes Sa2003, pp. 50–52) whereas was 16% for Calcutta (IOR/LG/AG.1). In Manila it was 25–30% for Canton or Amoy—for Spanish, Chinese or Danish Swedish borrowers alike (Van Dyke, Reference Van Dyke2005:154, Reference Van Dyke2011, pp. 44–45) but was 16% for Java (Ruiz Moreno, Reference Ruiz Moreno2023). These rates were significantly lower than what was paid for Acapulco ventures—in which silver was the return. It seems unlikely that premiums reflected the relative local availability of capital, but the expected returns given silver purchasing power in China, particularly.

Why would merchants with the same national/legal origin use dissimilar instruments in particular trades? Why merchants from different institutional backgrounds would use the same instrument in similar exchanges? Was a cashless system of payments necessarily a more “efficient” means of remittance when trade was in specie (gold/silver coins) as it was in the European trade with America and the East? Was the trade of silver a barter or an exchange transaction? As all cases of such trade share one common feature, namely, the contract stated the species of the return, the following pages will consider these monetary aspects.

4. Monetary nature of the bond for exchange

Financial historians consider silver as another commodity priced by a standardized weight and fineness—i.e. its intrinsic value (Esteves and Nogues, Reference Esteves, Nogues, Broadberry and Fukao2021); So, trading coined silver in foreign markets poses the question whether this was a barter of commodities priced according to supply and demand. Like cambium maritimum—as was known in the Mediterranean—the contract implied an exchange of currencies. In fact, by design, respondentia was an operation of change as mentioned in the introduction. Although premiums and principal were not always systematically distinguished in the contracts, in all cases there was a clear prescription of the monies or type of specie in the repayment. For example: “taes de prata de barras”; “Surat silver rupees of 30 heavy stuivers each,” “pesos Fuertes,” “piasters,” “pagodas at the exchange of 160 Spanish dollars per 100 star pagodas,” “after the rate of three taels in weight for the pound sterling” Footnote ²³ (see the Appendix); while pesos minted in Mexico was the default in Manila deeds. Historians of the Spanish Atlantic trade, highlight a feature of the contract that allowed a further “33% (extra) gain” on top of the premiums agreed, as the exchange involved specie as remittance back to Spain. (Bernal, Reference Bernal1992, pp. 318–19; Reference Bernal2013). Despite the obvious implications for the balance of payments, and the dissemination of silver beyond Spain, studies confined to national cases have overlooked this issue.

The role of foreign specie for mark-ups in Asia (and the potential for arbitrage) was apparent for the EIC Council at St. George in Madras as early as 1628:

The excerpt—one of many—showcases deals with foreign silver monies in India, and in Asia more generally, and the opportunities for arbitrage. The procurement of Chinese goods in Canton was a privilege of the Hong (guild of merchants) until the 1840s. Europeans advanced capital to them by using a combination of “at least 80%” silver specie and import goods, “or paid entirely of silver coins” (Van Dyke, Reference Van Dyke2011, p. 41 my emphasis). Prices clearly depended on the money of the advancement. Still in the 1800s, US merchant and Consul in Canton, Sullivan Dorr, reported that deals with silver coins would improve by 20% the price they ought to pay for teas in Canton. “Sometimes teas, silks and nankeens were not procurable on credit or by bartering with other goods … they were “cash goods.” Therefore, the correspondent advised: “loading ships only with (Spanish) dollars.” At the turn of the nineteenth century Dorr reckoned that “there was a premium on cash” for trade in Canton (Dorr, Reference Dorr1945:299 my emphasis). Still at the turn of the twentieth century, William Tate’ Modern Cambist reports that in China’s interior, principally in the tea districts where foreign (Mexican) dollars “formed a favorite medium of exchange,” they “circulated with a premium well above their intrinsic value” (Easton, Reference Easton1908, pp. 184–86). Hence, western nations coined their own “trade dollar” on the specifics of the Mexican coin to keep up trade with China between the 1870s and the 1900s (Irigoin et al., Reference Irigoin, Kobayashi and Chilosiforthcoming).

As foreign silver specie improved the terms of trade of Europeans in Asia, most of the deals involved cash. Asian economies performed with currencies of various metals, gold, silver, brass, or copper, which traded freely in markets and bazaars. Mughal India had free coinage and debasements were rare—at least through the seventeenth century—but recoinage was frequent. Multiple mints produced coins quite independently from each other, without coordination and with large differences in their output size.Footnote ²⁴ Thus, the rate of exchange between currencies was sensitive to price tradable goods and returns. The intense intra-Asian commerce increased the variety of foreign coins in circulation, further creating uncertainties about future prices for price-taking European merchants (Deyell, Reference Deyell and Richards1987; Haider, Reference Haider1996).

Paper instruments (hundis) long existed in Mughal India, issued by sarrafs—money changers/lenders—and spread to North and West India. These private bankers issued and discounted convertible hundis, offering a means to transfer money over distant places (Habib, Reference Habib1971; Haider, Reference Haider2019). The exchange rate was determined by the balance of payments among these markets, so sarrafs—who “specialised in respondentia” (Haider, Reference Haider1996, 299), regulated it by changing the rate accordingly. Factors of the EIC and the VOIC borrowed from sarrafs, too, as they were mostly remitters because their trade balance with India was largely negative. Europeans were thus also «takers» in the exchange rate (Habib, Reference Habib1971:294). Hundis worked fairly well as a means of payment and contributed to other means of settlement like book credit, although it is unclear to what extent they substituted coins in remittances within India (Habib, Reference Habib1971, 300).Footnote ²⁵ A study on the business of one of the largest Bengal bankers by mid-eighteenth century—and main lender to European Companies—shows that respondentia lending still generated more than a third of his earnings from financial assets, that included re-coinage, batta (currency exchange), tax farming, and other lending (Chaudhury, Reference Chaudhury2015, p. 91, table 1). Equally, merchants of Surat chose between bills and respondentia in lending to the English in the mid-1780s, according to the (current) “advantages of favourable exchange between Surat rupees and Spanish dollars needed for investment at Mokha.” (Nadri, Reference Nadri2008, p. 68).

Monetary diversity was especially high in South and South-East Asia, where sultans, rajas and princely states coined silver and copper. Foreign coins circulated by tale, and the exchange rate between coins fluctuated with private market prices. Arbitrage from regional differences in the price and exchange rates of silver coins was apparent for the EIC’ at Fort of St. Georges (Madras) in the mid-eighteenth century:

Thus, in a trade where differences between exchange rates and current cash maybe large, pricing and returns faced additional costs from currency risk.

In the sixteenth century, Mughal tried to standardize the silver in circulation with the sicca rupee, but this was not complete as it entirely depended on imported silver. Along with cowrie shells and other commodity monies in Gujarat and Bengal, for instance, foreign silver coins like Shahis from Iran, Larins from Hurmuz and Basra, and Persian Abassi circulated with a premium. Regional chieftains also had access to coinage, so Gujarat remained using their Mahmudi (88 grains of silver) and even issued inferior coins under Mughal names (between 75 and 87 grains) (Haider, Reference Haider1999, Reference Haider1996, p. 345). With European trade at Surat, by the mid-seventeenth century, the Spanish real—the piece of eight—became the preferred silver coin from Mokha to Canton. Increasingly from the 1630s, Asia imported from Europe silver, mostly coined in the shape of reals and pesos, while Dutch Ducatoons were relatively more significant in the East Indies. (Chaudhuri, Reference Chaudhuri1963, Reference Chaudhuri1968, table 1; Glamann, Reference Glamann1949; Dermigny, Reference Dermigny1964).Footnote ²⁷ This supply of foreign coins expanded together with the extraordinary growth in Spanish America silver coinage after the 1730s.Footnote ²⁸

Within the great variety in Asian monetary settings, the case of China stands out. China used units of account for both silver—i.e. the tael which was a weight measure 37.565 (grams) which varied locally and with the purpose, and for copper—i.e. (weng) a string of 1,000 coins of theoretically the same quality. Because they held a notional fixed (and invariant) ratio of 1 silver tael to 1,000 weng, historians have assumed a bimetallic monetary system in place. However, copper was a provincial and private coinage prone to debasement and counterfeits (king, Reference King1965; Vogel, Reference Vogel1987), so string collated any kind of coin.Footnote ²⁹ On the other hand, the empire never minted silver (or gold); thus China had no standard, mint parity or a par value, for silver. Indeed, China suffered a chronic shortage of means of payment of higher denomination, and from the mid-eighteenth century increasingly relied on the foreign coins coming to South China ports. “Silver (coins) was a public necessity in Asia”—and the “proof was in the price” (De Vries, Reference De Vries and Maxine2015, p. 24 my addition); and China had no bimetallic but a multiple monetary system.

This establishes a first difference of Asian monetary regimes with that of Europe; there gold and silver circulated by tale and sovereigns’ control over mintage was limitedFootnote ³⁰; neither money markets nor coinage was centralized, and they relied on imported silver for their own coinage or on foreign coins as means of payment in domestic transactions. Without official par or mint values, even without a mint where individuals could bring the silver, there was no way or standard to trade silver weight for silver weight (or gold) in China, but coin for coin, whichever the metal. Thus, high-quality, most abundant Spanish American coins enjoyed a growing premium over their content in China. As measured in sterling and recorded by the High Admiralty Courts in the Mediterranean: in the 1640s the coin was priced at 52–54d in Alicante, 54–57d in Malaga, 54–60d in Livorno, 56d Genova, between 48 and 57d in Venice and 57d to 66d in Smyrna; it was worth 48d in Cadiz and between 48d and 52d in London (Blakemore, Reference Blakemore2017). Still in the 1780s’ Asia the silver peso quoted at 62d in Basra, 63^1/2 in Bombay and 50d in Batavia, at 57d in Malacca and 64^1/2d in Surat (An account of monies, 1789)—a range that meant 28% difference between Batavia and Surat and a similar premium over the intrinsic value of the coin.

Monetary historians considering metallic currencies as commodity monies—units of metal weight by the fineness—owe a lot to models conceived in the bimetallism of Europe (Velde et al., Reference Velde, Weber and Wright1999; Flandreau, Reference Flandreau2004; for a nuance Bignon and Dutu, Reference Bignon and Dutu2014), but their assumptions do not help to appraise monetary developments in extra-European economies. It does not help to understand Spanish America’s either. The assumption that silver and gold as commodities were minted in “relatively small” volumes—as in Europe—does not hold considering that between 1772 and 1800 (when reliable population data are first available) Mexico alone minted 110 grams of silver of .916-thousands fine—i.e., about 4 pesos per capitaFootnote ³¹—every year of the world’s “dominant currency” and most current coin (Irigoin, Reference Irigoin and Battilossi2020).

Although Indian rulers also cut coins—mostly of silver—they sourced the silver as return from exports. Rupees had a varying market value according to the year and mint of issue and their fineness varied greatly, including the coinage at the three ports controlled by the Company (see appendix 2, The Madras Commercial ..., 1818, pp. 129–133; 1833, pp. 169–172). There was not even consistency among these rupees—and it was hard for the Company to keep the relation between their own gold pagoda with the gold mohur of Bengal within the gold/silver ratio in balance (Furber, Reference Furber1948, p. 350). Foreign coins could be converted into rupees at the imperial mints but in fact, the sarraf (money changers) established their value in open money markets. So, there was not a silver standard in India for foreign trade either until 1835, when a new Company rupee was based on the pound sterling.Footnote ³² In China, foreign coins were priced by private shroffs (money changers) too according to their assay of weight and touch (fineness).Footnote ³³ So, Europeans were also «takers» in the price of silver-until the Bogue (1843) and Tientsin (1858) treaties gave them a word on the rate at which foreign coins were converted in China (Irigoin et al., Reference Irigoin, Kobayashi and Chilosiforthcoming). Thus, instead of bimetallism in China, a complementary regime of metallic monies and a substitution of heterogeneous silver coins could happen simultaneously (Kuroda, Reference Kuroda2020, Reference Kuroda2008).

There were no “foreign coins” in Spain or Spanish America until the nineteenth century; and if at all, they were considered bullion. Until 1686, both metropolis and colonies performed with the same monetary unit with a similar name and denomination. Following wild debasement of copper (billon) in the 1640s, which brought considerable disorder to money markets, in the 1680s, Spain added a new lighter silver coin to the piece of eight—peso de 8 reals—minted in America. This debasement made the peso cut in Spain distinct from the “old” one minted in Spanish America; it was worth 10 silver reals or 15 reals 2 maravedis of billon, which depreciated the new coin in Spain by 33^1/3% to the old one. This gave way to a distinction of monies—physical and of account (or de cambio) in Spain and the colonies.

This separation of the monies in the metropolis and colonies ended with the specifications of gold, silver and monies of account—maravedis—established in the 1497’s royal Pragmática. Maravedis—an “imaginary moneyFootnote ³⁴”—was the basic unit of account; 34 of which made a real, and as eight reals made the peso, thus the (old) silver peso was worth 272 maravedis until 1686. In Spain, merchants and the Treasury used it in their accounts so everything there was valued as a compound of maravedis; but they were unknown in Spanish America where peso and reals maintained their intrinsic content steady—and remained the unit of account and only physical coin until the 1820s. While the Spanish Indies never issued small change of copper.

On the other hand, commercial transactions inside Spain (and with Europe) used gold and silver units of account and exchange that were not linked to a physical coin, e.g. the doblon for gold, and the ducado of 375 maravedisFootnote ³⁵ and, the peso of 128 cuartos of 512 maravedis for silver. The latter became the peso de cambio as a unit of exchange in bills. Its value declined relative to the base coin—of eight reals per peso (old silver) minted in America as more reals made a peso in Spain.Footnote ³⁶ This difference between the price of the coin and that of the money of account permeated in foreign trade; with England for instance, whereas the par exchange value of the peso de cambio oscillated around 40d, the mint par value of the coin was 54d—making too a 33% premium on the coin in London. This difference in the valuation of the exchange with the price of the coin repeated in the European money markets after the 1640s and should be considered in the research on silver flows in and out of each European economy of the time.

In these regards, it is interesting to recall two features of the cambium maritimum as was used in the thirteenth century Genoa according to De Roover (Reference De Roover1969), which fittingly apply to respondentia.Footnote ³⁷ First, De Roover says, the lender’s profit was determined by “undervaluing the foreign currency in which the loan was repayable” (p. 17) and “overvaluing it at the fair” (p. 20) or “determined by manipulating the exchange rates and setting arbitrary values” (p. 24). Second, being conditional on the ship’s arrival, the instrument “was rarely, if ever, used for the purpose of remitting funds to another country” (p. 23, my emphasis). Therefore, respondentia was indeed a means to bring in foreign money. In the eighteenth century, placing higher values on their own local currency than on the foreign one was only possible in some European countries—crucially this overvaluation drew silver money from the “first” to the “second receivers.”Footnote ³⁸

Europeans had struggled to establish a uniform coinage throughout the seventeenth century (Van der Wee, Reference Van der Wee, Coleman and Postan1977-2008), as they were caught between the need to import precious metals for coinage and the increasing demand for silver to trade in the Baltic, the Levant and Asia at large. They had, however, different means to manage the monetary system away from private agents; Europeans “standardized” money and “centralized” money markets (Neal, Reference Neal2002) from a medieval multidenominational currency minted in different metals.Footnote ³⁹ Along with their own coins, they developed monies of account (Spufford et al., Reference Spufford, Wilkinson and Tolley1986; Supple, Reference Supple1957)—i.e., a unit of measurement for metallic monies or an assigned quantity of weight of certain fineness like sterling—whose value tried to follow the market price of gold and silver, fixing the exchange rate in an «adjustable» number of their current monies. «Imaginary» monies (e.g. Venetian’ scudi d’or, the Genoese’ scudi de marche) developed as units of account that standardized payments and simplified the settlement of balances in foreign money in the exchange fairs. As a rule of thumb, current monies denominated in monies of account, tended to deliberately overvalue certain coins at home to avoid the melting of their own silver—already scarce—and to attract precious metals to their mints (Van der Wee, Reference Van der Wee, Coleman and Postan1977-2008: 297, 308).Footnote ⁴⁰

In the 1640s, the Amsterdamse Wisselbank’ managed the supply of bullion to Dutch mints and to money markets. It used bank money as money of account and centralized transactions with shorter and more regular turnover of bills, fostering a cashless system of payment (Quinn and Roberds, Reference Quinn and Roberds2012; Quinn and Roberds, Reference Quinn and Roberds2024). Thereby, public banks replaced the Genoese fairs, which traded in their own monies of account (Van der Wee, Reference Van der Wee, Coleman and Postan1977-2008, pp. 290–94, especially)Footnote ⁴¹—expanding the international system of transfer payments that matured in the eighteenth century (Nogues, Reference Nogues and Battilossi2020). Its bank guilder became a unit of exchange for international deals within Europe and both Dutch Indies, trading in precious metals, i.e., buying it from West Indies and Europe and selling it to mints, and for exports to the East (Vand Dillen, Reference Vand Dillen and Van Dillen1934; Guillard, Reference Guillard2004). Amsterdam became the chief money market in Europe and the marketplace for American silver (Morineau, Reference Morineau1986). Monies of account helped to moderate the swings in the value (quality) of foreign coinage prone to debasements or enhancements in the seventeenth century; this intermediation allowed the domestic economy to decouple from uncertainties in international markets. Other banks—subsequently in London and Hamburg, too offered a cashless means to settle balances in an overseas trade that coevolved with the financial development. This “ledger money” became quasi fiduciary (Quinn and Roberds, Reference Quinn and Roberds2024) and the agio Footnote ⁴² worked as a “sluice gate” (Neal, Reference Neal2002) that allowed a flexible exchange rate to fend it off from withdrawal of coins in reserve. In Amsterdam, the agio, the premium between bank money over the currency, oscillated around three and half and five percent between 1640 and 1775 (McCusker, Reference McCusker1978, pp. 43, 44, 62; Quinn and Roberds, Reference Quinn and Roberds2012, p. 20) while Hamburg’ bank offered 20–30% premium for their reichsthaler banco (Pfister, Reference Pfister2017). In 1761, for example, the agio was 25% in Genoa, 18% in Venice, 22% in Frankfurt am Main and Hamburg (Itinerario, 1761. p. 19, 29, 40, 43).

However, while exchanges were priced in bank money or monies of account, even in Europe, commodities and merchandize were transacted in current monies, and different commodities were priced in different currencies in the same market (Posthumus, Reference Posthumus1946, vol. LX; McCusker, Reference McCusker1978, p. 44 for Amsterdam, p. 62 for Hamburg). In the West Indies, Europeans used the monies of account of their metropoles, but colonists set their own currencies on the most current Spanish American silver coin (piece of eight standard) and used it as numeraire and unit of account. Even French and English merchants in the Caribbean negotiated among themselves on the Spanish American peso (McCusker, Reference McCusker1978: 287). According to Denzel (Reference Denzel2010, p. xlvii), bills were ultimately “credit instruments for metropolitan merchants” only. In North America, local assemblies set and repeatedly “cried up” their monies value—and did it separately—so exchange rates in sterling diverged wildly (McCusker, Reference McCusker1978, pp. 125–229, 234–256; Mossman, Reference Mossman1993, pp. 48–50, 62–63). Eventually Queen Anne’ Act of 1704 (Parliament’ Act of 1707) undervalued the Spanish coin in the colonies at 72s—the “Proclamation Money”—a third below the mint par value the peso coin had in Britain (54d).Footnote ⁴³ So did the French in their Caribbean colonies—and for the same 33 ½% (McCusker, Reference McCusker1978, p. 282) and in Canadian settlements (Chalmers, Reference Chalmers1893, p. 14); in Brazil the Portuguese “cried up” their current money too by 33% and more during the gold boom and the renewal of minting there in the 1710s (Cerqueira Lima and Martins De Souza, Reference Cerqueira Lima and Martins De Souza2017).Footnote ⁴⁴ Even English coins were valued higher in the colonies: the silver crown was worth 60 pence in England, but it was rated at 75d by the Virginia Act of 1727 (Chalmers, Reference Chalmers1893, p. 13), 90d in 1759 Philadelphia and 96d in New York (Mossman, Reference Mossman1993, table, p. 75) and 81 in Massachusetts.Footnote ⁴⁵

Monies of account were unknown in the Spanish Indies, her peso coin—called dollar in English in lieu of the (real) after the 1680s changes—was the currency and unit of account that served to price goods and exchange. As the coin enjoyed a steady intrinsic value throughout (albeit three minor debasements in 1732, 1772, and 1786), variations in the exchange were driven by devaluations elsewhere or changes (enhancements) in metropolitan monies of account. It was Spain—with a much less centralized coinage—that in the later eighteenth century counted twenty different types of “physical current” coins of silver, gold, and copper and various «imaginary currencies»—or monies of account—(Larruga y Boneta, Reference Larruga Y Boneta1787 vol. I, p. 73) Spain was the only one of the European powers with an undervalued domestic rate of exchange with her colonies currency.

The European companies in East Indies performed mostly with coins; the VOC largely shipped their own coins—even the small change—to Java and Batavia (Feenstra, Reference Feenstra2020)Footnote ⁴⁶—whereas the English, who coined nearly no silver at home and run three mints in India, consistently re-exported foreign silver coins to Asia since the mid seventeenth century (Chaudhuri, Reference Chaudhuri1963, Reference Chaudhuri1968). Behind this difference, however, both companies alike undervalued the foreign coin in Asia to procure silver there and reduced shipments from Europe.Footnote ⁴⁷ From 1664 to 1734, the Dutch company issued assignaties for 237 million florins—i.e., 4% interest-bearing bills (Pol, Reference Pol1985, p. 90). They were denominated in monies that priced lower than the silver ducatoon in Asia—e.g., at 78 stivers in Batavia (it went to 72 between 1732 and 1782) against 63 in Amsterdam (Gaastra, Reference Gaastra1994). These bills helped mobilize private capital to finance further imports of Asian goods, although they did not fully substitute for the export of silver (Gaastra, Reference Gaastra1994, fn5). This also afforded a 24% profit to private agents in the exchange rate difference between Batavia and Holland (Pol, Reference Pol1985, p. 141). Historians have thus underestimated the silver share of Dutch exports—particularly from within Asia—because coins were reckoned at a premium in money of account (Haider, Reference Haider1996, p. 321, Prakash, Reference Prakash1998/2008, pp. 88–89).

Having traded the Spanish American silver in Asia at 54d since 1619, over a mint par in London of 50d, from 1681 the East India Company invoiced the dollar coin at 60d per ounce sterling in their bills in Asia, adding transaction and transportation costs to the coins prime cost in London—at 54d after 1720 (Morse, Reference Morse1926–29, vol. I, p. 47). In the 1770s, EIC bills on Canton denominated in Spanish dollars now “at the exchange of the season”—a fixed but «adjustable» exchange rate—sought to procure specie in Asia for the China trade. Competition pushed up the exchange rates: the EIC further increased it to 62d and 66d, but the Danish and Swedish were offering both higher rates and shorter sight (Glamann, Reference Glamann1949). In 1787, the EIC started issuing 30-day sight bills on Bengal at 39 rupees per 100 dollars, a premium over the current exchange par of 41 ½ (Morse, Reference Morse1926–29, vol. II, p. 142). Bills in Spanish dollars went up to 66–68d after 1797’ suspension of the pound convertibility in England and during the French Wars jumped to 84d; the rate for rupee bills climbed to 42–43 per 100 dollars (Morse, Reference Morse1926–29, vol. II, pp. 358, 388) and paid up to 69–66d in 1801–02 and 1804–05 in London for the coins (Report, 1810, table 13, 168). There were even no purchases in 1809 and in 1810–11, extraordinarily, flows reverted, i.e. London imported silver dollars from China and India.

As in the Caribbean, in Asia the dollar/peso coin was too the “measure of all money, [the] universal basis for the exchange quotations (McCusker, Reference McCusker1978, pp. 280, 281).” However, the exchange par was determined by the purchasing power in private markets «free of regulations». As pointed out by Quinn and Roberds, the market price of coins commonly exceeded their legal value (Quinn and Roberds, Reference Quinn, Roberds, Neal and Atack2009, p. 36).Footnote ⁴⁸ Thus, neither in the East or in West Indies, the ordinance value on the piece of eight/dollar did ever reflect the actual par of exchange at the commercial rate and far less the intrinsic value. As Roover pointed “the manipulation of exchange rates and setting arbitrary values” to coins at home and abroad, conceived in the cambium maritimum—and the respondentia—ultimately affected the flow of silver. Thus, bills did not circulate outside the companies (and without the bill of lading), other than as remittance of profits or for coins procured throughout Southeast Asia rather than of credit, and at a «certain» exchange rate.Footnote ⁴⁹

Contrarily, respondentia contracts thrived in all European deals with China in Southeast Asia—at the peak of the trade in the 1780s (Van Dyke, Reference Van Dyke2011, pp. 41–48). Private bills appeared only much later in Asia when the standard of the silver dollars was collapsing. Sound coins were short, so the bills rate was at a more favorable rate from 65–66s in the 1800s, it went down to 51d in 1825 and further to 46d in 1828 (Greenberg, Reference Greenberg1951/1969, 160) when Chinese imports of silver nearly abated. They appeared in the Canton trade in the 1820s, mostly via US merchants drawing on London and British houses in India (Irigoin et al., Reference Irigoin, Kobayashi and Chilosiforthcoming).Footnote ⁵⁰

As late as 1832, Mr. Palmer, then governor of Bank of England and partner in the Calcutta house of Palmers, Mackillop and Co, declared to Parliament:

Thus, in long-distance commerce bills could not be an efficient cash substitute beyond Europe, without some monetary authority or institutions capable to set and regulate the value to the foreign coins. In most of Europe, foreign silver coins were received by count and were priced in monies of account that tended to overvalue them. In Britain, however, they were priced by weight (e.g., in ounces) (Tate 1908, p. 318) at the ratio of the sterling standard—i.e., the mint par established in 1604 and restored in 1702 by Newton assay.Footnote ⁵¹ By 1779 in Canton, the Spanish dollar had already become current and had “acquired an imaginary value through convenience”; the Dutch were already “pass(ing) the silver coin by tale instead of weight” (Morse, Reference Morse1926–29, p. 280; Van Dyke, Reference Van Dyke2011, p. 43). Chinese assayers were pricing exports and imports of silver, including gold coins, in Spanish dollars. In 1796, the EIC followed suit. By then, the species was already trading at 9% premium over its intrinsic value (Morse, Reference Morse1926–29 II, pp. 41, 313). Still in the mid-1830s, the Canton General Price Current quoted prices of exports and imports in Spanish dollars. Within Europe, this monetary anchor enabled bills’ substitution of precious metals for a cashless means of payments—when prices fell within import/export bullion points) (Sperling, Reference Sperling1962; Irigoin et al., Reference Irigoin, Kobayashi and Chilosiforthcoming for the data). This was impossible in Asia, where cash had a premium determined by the demand—which moreover was increasing after the 1790s. Same as in America, the source of the cash had an overvalued currency driven by the very large Asian demand. In this context, Roberds’ argument for an exceptional monetary capacity of Europeans is more persuasive than those grounded on her political institutions (Roberds, Reference Roberds2016: 219; Karaman et al., Reference Karaman, Pamuk and Karaman2020).Footnote ⁵² Thus, Europeans chose respondentia to settle exports to the New World and to pay for their imports in Asia, while using bills in more “certain” money markets inside Europe only—or in exchanges within a “more convenient” rate the Companies occasionally offered, as Governor Palmer attested above.