- 1 M. A. Denzel, 2015.
- 2 M. A. Denzel, 2020, pp. 283-289.
- 3 F. L. Hoffmann, 1849.
- 4 B. Backe-Dietrich & E. Lembcke, 2003.
- 5 B. Backe-Dietrich & E. Lembcke, 2003.
- 6 Katalog…, 1864.
- 7 Katalog…, 1864, col. 401-696.
1Strategically located on the River Elbe, about 100 kilometres inland from the North Sea, Hamburg has played an important role in much of Europe’s international commerce since the early modern period.1 In the course of the eighteenth century, Hamburg developed into a significant marine insurance market.2 The Hamburg merchant community leveraged the accumulation of commercial knowledge through the establishment in 1735 of a commercial library (Ger. Kommerzbibliothek), the world’s oldest private economic library.3 The goal of this library was to collect the relevant German and foreign literature for the education of wholesale merchants and to make it available to all for everyday use. It aimed to support knowledge acquisition and distribution within the Hamburg merchant community and, in turn, to promote Hamburg’s international commerce.4 Community members subsidized the purchase of relevant handbooks and reference works on the conduct of international trade. The library also possessed a collection of domestic and foreign weights and measures. Even though most of the Commercial Library’s collections were lost during World War II and the 1962 flood, and are now reduced to a fraction of their original size, descriptive catalogues provide ample insight into the kind of knowledge made available through these collections to Hamburg’s merchant community. Alongside maps, travel accounts, works on economics, state governance, nautical sciences and construction sciences, the “science of trade” (Ger. Handelswissenschaften) formed a significant part of the library’s collections.5 In 1864, when a new catalogue was published,6 the library comprised a wide variety of works ranging from trade journals, dictionaries, encyclopaedias, textbooks and general works on the history of trade, to reference works on metrology, bookkeeping, merchant correspondence and numismatics. Moreover, substantial sections dealing with commercial, maritime, exchange and insurance law were included under the Handelswissenschaften heading.7 A small subdivision of the section on commercial accounting (Ger. Kaufmännisches Rechnungswesen) lists works that contain commodity calculations or Waarenberechnungen.
- 8 Katalog…, 1864, col. 505-507.
- 9 E.g., J. M. Schwarzer, 1771; C. G. E. Krüger, 1791.
- 10 E.g., A. F. W. Crome, 1787.
- 11 [J. A. Engelbrecht], 1772, I; id., 1772, II; id., 1773, III; id., 1774, IV.
- 12 J. A. Engelbrecht, 1782, vol. 1, p. [1-2].
- 13 H. Schröder, 1854, II, p. 190.
- 14 H. Stettner, 1993, p. 81. For a brief overview of Engelbrecht’s works, see: H. Schröder, 1854, II, (...)
2Commodity calculations (hereafter: CC) are concise summaries of the costs involved in purchasing or selling a certain amount of a commodity on the international market. CCs describe the various payments to be made at the port of origin of the cargo, during shipment and at the final destination. They also provide information about measurement and currency conversions. CCs gave merchants a typical example calculation for estimating profitability under an outlined set of circumstances. Admittedly, not all items in the 1864 library catalogue fit equally well into the list of works comprising CCs. Excluding tables for calculating commodity prices in different locations, the subsection contains bibliographical references to nine different collections of CCs, published in English and German.8 CCs were often printed in accounting handbooks for merchants,9 or in specialist works dealing with bills of exchange.10 One large collection of 381 CCs, entitled Hamburgische Waarenberechnungen oder Sammlung richtiger und ausführlicher Calculationen verschiedener von andern Handels-Plätzen nach Hamburg gesandter, oder von Hamburg nach anderen Orten verschickter Waaren [Eng. Hamburg Commodity Calculations or Collection of Correct and Detailed Calculations of Commodities Sent from Different Trading Places to Hamburg, or from Hamburg to Other Locations], was published in four volumes by Johann Philipp Christian Reuss between 1772 and 1774.11 The collection’s compiler was not named. Only the introduction to the second, improved edition contains proof that Johann Andreas Engelbrecht (1733–1803) was responsible for compiling the CCs first published in 1772–1774.12 Born and raised in Hamburg, Engelbrecht moved to Bremen in 1756 to take charge of the English and French correspondence of several merchant houses.13 Later, he specialized in the insurance business. Throughout his career, Engelbrecht was a prolific translator of English, French and Dutch handbooks on marine law, insurance and trade, but also of popular literary works.14 In the following, we will refer to the first edition of the CCs as the Engelbrecht collection.
- 15 J. Hoock, P. Jeannin & W. Kaiser, 1991-2001; M. A. Denzel, J.-C. Hocquet & H. Witthöft, 2002; D. B (...)
- 16 M. A. Denzel, 2002, pp. 11-45; D. Besomi, 2012.
3Historians have paid significant attention to early modern merchant manuals, economic dictionaries and encyclopaedias. Major bibliographical, historiographical and lexicographical analyses of early modern economic literature have been published since the 1990s.15 The volumes dealing with commodity calculations seem to be largely forgotten, however, although their publication was part of the same trend aiming to provide an increasingly detailed description of the “science of trade” in the specialist literature. This process went hand in hand with the emerging standardization of commercial practices and the dissemination of commercial knowledge in printed manuals, dictionaries, encyclopaedias and practical reference works such as Engelbrecht’s Hamburgische Waarenberechnungen.16
- 17 J. L. Van Zanden & M. Van Tielhof, 2009, pp. 389-403; M. Van Tielhof & J. L. Van Zanden, 2011, pp. (...)
- 18 E.g., M. Van Tielhof, 2002, pp. 255-290; C. Deggim, 2005.
- 19 G. Federico & A. Tena-Junguito, 2019; S. I. S. Mohammed & J. G. Williamson, 2004.
4Due to its very nature, the Engelbrecht collection has the potential to shed new light on the historiographical issue of the value and meaning of transport costs and freight prices for research in economic history. The detailed calculations of transport costs that appear in the CCs differ in three important ways from the transport costs and freight prices that historians have used to study productivity changes in early modern shipping, or the question of an “early modern” transport revolution.17 The first, and perhaps most important, difference is that the various constituent parts of transport costs, from ship loading and unloading to packaging and storage at the port, are omitted from most economic-historical studies of transport costs and freight prices. Only a handful of available studies offer insights about the different logistical services involved in the early modern commodity trade.18 Compared to existing databases of historical freight rates, which all start well into the nineteenth century,19 the CCs offer a more encompassing view of transport costs. Freight prices are important, but they are only part of a long list of cargo transport costs incurred at origin, during shipment and at destination.
- 20 M. Van Tielhof & J. L. Van Zanden, 2011, p. 62ff.
- 21 M. Van Tielhof & J. L. Van Zanden, 2011, pp. 58-59.
- 22 M. Van Tielhof & J. L. Van Zanden, 2011, p. 47.
- 23 M. Van Tielhof & J. L. Van Zanden, 2011, pp. 57-58.
- 24 M. Van Tielhof & J. L. Van Zanden, 2011, p. 58.
- 25 M. Van Tielhof & J. L. Van Zanden, 2011, p. 64.
- 26 [J. A. Engelbrecht], 1772, I, p. 8-9.
5The second difference is that the Hamburgische Waarenberechnungen do not contain a homogenous series in the way that, for example, Dutch freight rates do. This becomes particularly clear when we compare the contents of the CCs with the work of Van Tielhof and Van Zanden on freight rates, in particular their 2011 book chapter Productivity Changes in Shipping in the Dutch Republic: The Evidence from Freight Rates. Relying on previous historiography and archival research, Van Tielhof and Van Zanden created an impressive dataset of more than 2,800 different freight prices for Dutch shipping on the main European routes from Amsterdam in the early modern period. Scattered evidence of freight prices was collected for ships chartered on the routes between Amsterdam and Archangelsk, the Baltic, Bordeaux and the Mediterranean.20 Most of the freight prices included in their work were taken originally from charter-parties found in the Amsterdam notarial archives. In these contracts, freighters or charterers and shipmasters agreed upon a certain price for a specified transport service.21 In their chapter, Van Tielhof and Van Zanden define freight rates as “prices of output”, which are contrasted with the cost of shipping and shipbuilding (“prices of input”) to determine changes in total factor productivity in early modern Dutch shipping.22 To overcome the limitations of previous research that often relied on nominal freight rates, the method for estimating total factor productivity in Dutch shipping requires a long series of deflated freight prices.23 In this respect, the authors’ method fundamentally differs from earlier, more straightforward calculations of “freight factors”, i.e., the share of the freight rate in the total price of the commodity transaction.24 In the Hamburg case, freight rates were not “usually expressed in guilders per [rye] last,”25 nor is it possible to convert the given freight prices in the CCs to one common denominator because the prices described as Fracht (freight) in the CCs and the Dutch freight prices, also referred to as Fracht or, in Dutch, vracht, do not have the same meaning. The Dutch “freight prices” indicate the cost of chartering an entire ship, whereas the Fracht in the CCs corresponds to the price charged for using available storage space on a (chartered) ship to forward a quantity of a certain commodity. For this reason, the freight prices in the Hamburgische Waarenberechnungen were calculated differently. In contrast to the Dutch prices in guilders per last, indicative of the price of chartering an entire ship, the freight prices in the CCs are given as a percentage of the value of the cargo, as a negotiated value (accordirt), or as a fixed rate per measure (e.g., per sack). As a result, different freight prices are specified for the various legs of a voyage described in the CCs. For example, for a cargo of anchovy destined for Danzig, one freight price is given in Mark banco to cover the distance from Hamburg to Lübeck, and another freight price in Polish florin to cover the seagoing leg of the voyage from Lübeck to Danzig.26
- 27 M. Van Tielhof & J. L. Van Zanden, 2011, p. 65.
- 28 M. Van Tielhof & J. L. Van Zanden, 2011, p. 65.
- 29 In his economic encyclopaedia, Krünitz provides a list of corresponding French terms for Caplaken, (...)
- 30 J. G. Krünitz, 1773–1858 [online], lemma Kapplaken.
6The third difference is that in most studies based on freight rates “the particular conditions and arrangements made in the contracts, such as […] the hat-money or special reward for the shipmaster personally (Nl. kaplaken)” are not taken into account.27 It is true that their relevance to transport costs in general was limited and that they did not tend to diminish or increase over time.28 However, additional cost items such as Caplaken and primage were integral parts of the shipmasters’ remuneration and should therefore be included in the freight prices. Caplaken, hat-money or (droit de) chapeau,29 was a sum of money that the a ship’s captain received on top of the agreed freight price and should be understood as an additional incentive to take good care of the cargo. Primage, Prämie or Priemgeld is a related concept that emerged when Caplaken started to become an integral part of the freight price in the course of the eighteenth century, resulting in the need to devise an alternative incentive.30 In practice, freight, Caplaken and primage often appeared next to each other.
7The next section contains a more detailed introduction to the Engelbrecht collection, clarifying its relevance for historiography and explaining how selected CCs were analysed. The section on cost items adds more depth to the survey and examines the composition of the various cost items as found in the CCs. Different cost items in Hamburg and at the ports of origin or destination of the goods are examined and their share in the total transportation cost for entries in the compilation is estimated.
8The Engelbrecht collection constitutes the basis for this paper. Oddly, neither edition was found in the 1864 catalogue of the Hamburg library of commerce. However, similar bibliographical reference works of that time, such as Palm (1790) and Gruber (1794) did list Engelbrecht’s collection of Hamburgische Waarenberechnungen, and praised its comprehensiveness and accessible format.31 Johann Christian Philipp, a merchant and trader in Dresden, went further and referred explicitly to this collection as source of inspiration for his own work.32 Moreover, some excerpts found their way into Johann Georg Krünitz’ famous Ökonomische Enzyklopädie oder allgemeines System der Staats-, Stadt-, Haus- und Landwirthschaft, in alphabetischer Ordnung [Eng. Economic Encyclopedia or General System of the Economies of State, City and Household and of Agriculture, in Alphabetical Order].33 Such evidence of its distribution and practical use confirms that the Engelbrecht collection must have been a respected source within and beyond the Hamburg merchant community. As part of VD18 digital,34 a recent funding initiative for the digitization of printed works published in German in the eighteenth century, the four volumes of the Engelbrecht collection have been scanned and published online by the Rostock University Library.35
- 36 [J. A. Engelbrecht], 1772, I, p. VII; id., 1772, II, p. III-IV.
9The CCs in the Engelbrecht collection are between one and three pages long. Although the observation period remains implicit in the source, the prefaces to the different volumes provide some helpful indications. In volume one, Engelbrecht states that he has been collecting CCs for several years prior to publication, and that, since announcing the first volume, he has continued to receive additional materials from members of the Hamburg merchant community.36 Moreover, Engelbrecht points out that most of the CCs only contain facts and very few have been handed over to him as conti sinti, i.e., fictitious calculations (Ger. fingierte Calculationen). This is important, since it separates this collection from the more common use of fictitious calculations in merchant manuals. At the same time, users should be aware that Engelbrecht provides no information about the merchants who have made the purchases and then sent him their CCs, making it is almost impossible to verify and compare the calculations with entries in real account books.
10The Engelbrecht collection contains CCs for about 120 different commodities, and as such is highly representative of Hamburg’s international trade in the second half of the eighteenth century. Only for a few commodities were more than four CCs included in the collection. These commodities (and their frequency in brackets) are almonds (5), brandy (6), coffee (20), cotton (14), currants (6), grain (6), ginger (5), indigo (9), leather goods (6), oil (12), raisins (5), syrup (5), sugar (13) and wine (7). The majority of the CCs describe goods arriving in Hamburg (289 out of 381 CCs), and reflect the port’s main trading partners in the second half of the eighteenth century (Table 1).
Table 1. Geographical distribution of CCs describing arrivals at Hamburg
Country |
|
Predominant ports (number of CCs in brackets) |
Great Britain |
69 |
London (44), Liverpool (8), Bristol (7) |
France |
56 |
Marseille (19), Bordeaux (9), Lorient (7), Nantes (6) |
Dutch Republic |
45 |
Amsterdam (40) |
Italy |
44 |
Livorno (23), Venice (10) |
Spain/Portugal |
34 |
Malaga (10), Alicante (8), Lisbon (6) |
Other |
41 |
Lübeck (6), Copenhagen (6) |
Source. [J. A. Engelbrecht], 1772-1774.
11Besides Marseille, Bordeaux, Lorient and Nantes, 12 other French ports are present in the compilation. These ports are (in alphabetical order with number of CCs in brackets): Bayonne (4), Blaye (1), Brest (1), Sète (1), Le Croisic (1), Dunkerque (1), Le Havre de Grâce (2), Montpellier (2), Paris (1), La Rochelle (5), Saint-Malo (1) and the coastal area Charente (1). The destinations of CCs for goods exported from Hamburg underline the port’s special role as a gateway between Western and Southern Europe, on the one hand, and the Baltic Sea region, on the other. Indeed, the most frequent destinations were Danzig (26), St. Petersburg (7), Lübeck (6), Königsberg (5) and Riga (2) which together account for almost half of all different destinations found in the Engelbrecht collection.
- 37 [J. A. Engelbrecht], 1773, IV, pp. 100-101.
- 38 [J. A. Engelbrecht], 1773, III, pp. 7-8.
- 39 One special CC, documenting the purchase and delivery of cotton from Smyrna to Marseille, is not c (...)
- 40 [J. A. Engelbrecht], 1772, I, pp. 55-56, 127-128, 148-151; id., 1772, II, pp. 134-135, 142-143, 17 (...)
12With regard to the geographical distribution of the CCs in the Engelbrecht collection, three further aspects are worth stressing. First, a number of inland trading centres, such as Leipzig, Rovereto (spelled as Roveredo in the collection),37 Nuremberg, Lille (Ryssel in the collection),38 and Paris, are represented. Second, 23 CCs contain explicit information about intermediate stops between departure and destination. These CCs underline the importance of the route via Lübeck as a means to avoid the Danish Sound dues, while also shedding light on Hamburg’s role as destination of trans-Alpine trade routes from Venice and Trieste via Nuremberg. Moreover, some CCs provide evidence of the role of ports in the Mediterranean and the Atlantic, such as Livorno, Marseille, Sète or Cadix, for example, that served as entrepôts where goods were held prior to onward shipment. Third, 14 CCs describe calculations for commodities that were purchased on behalf of Hamburg merchants, but never actually went through the port.39 Transactions of this type occurred in both directions and often involved some of Europe’s major ports.40
- 41 [J. A. Engelbrecht], 1772, I, pp. 7-9.
- 42 [J. A. Engelbrecht], 1772, I, pp. 20-27; id., 1772, II, pp. 11-12.
- 43 [J. A. Engelbrecht], 1772, I, p. 37; id., 1772, II, p. 13; id., 1774, IV, p. 5.
- 44 [J. A. Engelbrecht], 1772, I, pp. 53-54, 57-58, 61-64, 66-67, 74-76; id., 1772, II, pp. 18-19, 24- (...)
- 45 [J. A. Engelbrecht], 1772, I, pp. 81-84.
- 46 [J. A. Engelbrecht], 1772, II, pp. 81-84; id., 1773, III, pp. 45-46.
- 47 [J. A. Engelbrecht], 1773, III, pp. 125-126.
- 48 [J. A. Engelbrecht], 1772, II, pp. 181-184; id., 1773, III, pp. 178-180, 183-186; id., 1774, IV, p (...)
13Processing and using the Hamburgische Waarenberechnungen turned out to be quite challenging. Given the complex contents of the source and the limitations of the first digitization, which provides scans but not a machine-readable version of the source, it was only possible to focus on a selection of entries from the compilation rather than trying to deal with all CCs at once. Naturally, therefore, the insights presented in this paper are of a preliminary nature. From the four volumes of the Engelbrecht collection, we selected a number of entries with a direct relationship to France, i.e. commodities leaving or entering French ports, since they are well represented in the CCs. To facilitate comparison, admittedly on a limited basis, additional commodity calculations for the same goods as the French entries were also included. The result is a dataset of 33 CCs corresponding to almost 9 percent of the total number of calculations in the Engelbrecht collection. It contains CCs for anchovy,41 cotton,42 (iron) sheets,43 coffee from Martinique,44 capers,45 liquorice juice,46 staves,47 and sugar from Saint Domingue.48 The selected entries provide a first opportunity to place freight rates and transport costs in the context of all costs incurred in the purchase or sale of commodities on the international market. To some extent, they also provide a means to compare costs for the same commodity from different locations, to look for differences in the cost of transporting goods from the same location, and to interpret different freight prices and other costs. Table 2 provides a translated example of one CC for cotton imported from Bordeaux. The line numbers in the following descriptive account of this example refer to the rows in the table and correspond to the lines of printed text on the scans of the source.
Table 2. Cotton from Bordeaux (“Baumwolle von Bourdeaux”)
Page |
Part |
Line |
Cotton from Bordeaux |
22 |
A |
1 |
There were bought 12 bales of Smyrna cotton, which upon Receipt |
|
|
2 |
weighed Brutto 2909 ℔ |
|
|
3 |
Tara at 4 pC. 116 [℔] |
|
|
4 |
― |
|
|
5 |
net. 2793 ℔ - at 55 Liv[res] for 100 ℔ £ 1536=3=— |
|
B |
6 |
Costs: |
|
|
7 |
Export toll on 2700 ℔ - at 5 £ - 135=—=— |
|
|
8 |
2 months storage rent a 10 S[ous]. per bale |
|
|
9 |
(caused by a lack of shipping capacity when the |
|
|
10 |
purchase was made) - - 12=—=| |
|
|
11 |
Ropes - a 25 S[ous]. - - 15=—=— |
|
|
12 |
Loading 12 S[ous]. - 7=4=— |
|
|
13 |
Courtage ½ pC. - - 7=13=— |
|
|
14 |
― |
|
|
15 |
176=17=— |
|
|
16 |
― |
|
|
17 |
£ 1713=—=— |
|
|
18 |
Provision 2 pC. 34= 5=— |
|
|
19 |
― |
|
|
20 |
£ 1,747= 5=— |
|
|
21 |
exchanged at 27¾ [Schilling] pr[o]. Krone [Mark banco] 1,010= 3=— |
|
C |
22 |
Local costs: |
|
|
23 |
Insurance on 1,050 [Mark] - at 1½ pC. - [Mark banco] 15=12=| |
|
|
24 |
Customs 6 [Schilling] 6 [Schilling] 5⅓ [Schilling] - - - 13—=— |
|
|
25 |
Convoy at ½ pC. of [Mark] 1,000 - [Mark] 5=—=| |
|
|
26 |
Freight at 9 [Mark] pr[o]. Bael [Mark] 108=—=— |
|
|
27 |
Hat-money [Caplaeken] at 1 [Mark] - 12=—=— |
|
|
28 |
Average [Averie ord.] at 12 [Schilling] p[ro]. Bael 9=—=— |
|
|
29 |
Customs at Stade - 3=12=— |
|
|
30 |
― |
|
|
31 |
132=12=— |
|
|
32 |
Unloading, weighing |
|
|
33 |
and delivering - at 10 [Schilling] - 7= 8=— |
|
|
34 |
Courtage on 1,260 [Mark] ⅚ pC. - 10= 8=— |
|
|
35 |
Same of insurance on 1,050 [Mark] ¼ pC. 2=10=— |
|
|
36 |
Storage rent and letter postage 6=—=— |
23 |
|
37 |
Transport - [Mark courant] 164= 6=| [Mark banco] 28=12=| [Mark banco] 1010= 3=| |
|
|
38 |
Agio 20 pC. 27= 6=— |
|
|
39 |
― |
|
|
40 |
137=—=— |
|
|
41 |
― |
|
|
42 |
165=12=— |
|
|
43 |
― |
|
D |
44 |
These 12 bales weighed here 2,980 ℔ upon receipt [Mark banco] 1,175=15=— |
|
|
45 |
Allowance [Gutgewicht] 1 pC. 30 - |
|
|
46 |
― |
|
|
47 |
2,950 ℔ |
|
|
48 |
Tara at 4 pC. 118 |
|
|
49 |
― |
|
|
50 |
2,832 ℔ - 14 7/16 ₰ - [Mark] 1,277=11= 6 |
|
|
51 |
13 month discount 101=14= 6 |
|
|
52 |
― |
|
|
53 |
[Mark banco] 1,175=13=— |
Note. Example translation of one commodity calculation taken from [J. A. Engelbrecht], 1772, I, pp. 22-23.
- 49 [J. A. Engelbrecht], 1772, I, pp. 22-23.
- 50 M. A. Denzel, 2010, p. 191.
- 51 M. A. Denzel, 2010, pp. 191-192.
- 52 J. Schneider, O.-E. Krawehl & M. A. Denzel, 2001.
- 53 M. A. Denzel, 2010, p. 192.
14The “cotton from Bordeaux”49 CC (Table 2, line 1) contains details about 12 bales of cotton from Smyrna that were bought in Bordeaux and had a gross weight of 2,909 pounds when received (line 1-2). From its gross weight, four percent tara (weight of the commodity container or package) were deducted (line 3), resulting in a net weight of 2,793 pounds (line 5). The cotton was valued at 55 livres per 100 pounds or 1,536 livres and 3 sous in total (line 5). In Bordeaux, several expenses were incurred (line 6-13). Export taxes were paid at a rate of 5 livres per 100 pounds (line 7). Storage space for two months was bought, due to the lack of available shipping capacity at the time of the purchase (lines 8-10). The rent was set at 10 sous per bale (line 8). Packaging materials for the bales were bought at the price of 25 sous per bale (line 11). Loading and storing the bales cost 12 sous per bale (line 12). Brokerage of 0.5 percent on the value of the cargo completed the expenses incurred at the port of Bordeaux (line 13). The sum of these expenses, 176 livres and 17 sous (line 15) was added to the value of the purchase, raising it to 1,713 livres (line 17). To this value, a two-percent commission was added (line 18), bringing the total value to 1,747 livres and 5 sous (line 20). This value was converted to Mark banco with an exchange rate of 27 ¾ Schilling (line 21). The Mark banco was the “never-minted currency of account of the Hamburg banco, which was established in 1619 […] to stabilize the Hamburg currency and exchange”.50 The corresponding total value of the purchase in Mark banco was 1,010 Mark and 3 Schilling. In Hamburg, several further costs (hiesige Unkosten, Table 2, line 22) were added, some calculated in Mark banco (line 23-24), but most often expressed in Mark or Mark courant (lines 25-29 and 32-36). The Mark courant was the “city’s real means of payment”, i.e. a minted coin that was quoted with a significant discount, or agio, against bank money.51 As a rule, the cost items given in Mark banco were insurance and tax payments (line 24). The insurance rate was 1.5 percent (line 23). Although this example does not specify names for the different taxes due in Mark banco (line 24), we can see that three different rates (in Schilling) were given, corresponding to the burgher, civic and admiralty tolls (Ger. Herren-Zoll, Bürger-Zoll and Admiralitäts-Zoll). Together with the 0.5 percent convoy tax (line 25), which was calculated in Mark courant, the admiralty tax is well-known in the historiography of Hamburg’s trade.52 In lines 26-28 of the CC freight prices, Caplaken and Averie ordinaire were calculated based on a fixed rate per bale of cotton. Nine Mark per bale were due for the freight price (line 26), one Mark per bale for Caplaken (line 27) and 12 Schilling per bale for Averie ordinaire (line 28). Another customs payment, the tax at Stade in the mouth of the River Elbe, was due in Mark courant as well (line 29). Following an intermediate sum of the cost items in lines 26-29, more expenses were listed that had to do with the cargo itself (lines 32-33), its storage (line 36) and brokerage fees (lines 34-35). Continuing on page 23, the key figures of the calculation so far were carried over from the previous page (line 37). These data inform us that the total cost in Mark courant was 164 Mark and 6 Schilling, i.e. 137 Mark banco (line 40). From the cost in Mark courant, a 20 percent agio was discounted (line 38) to align this value with the expenses incurred in Mark banco. The 20 percent discount seems to be a rough, but not unrealistic estimate of the average annual discount of the Hamburg Mark courant against the Mark banco around the time the Engelbrecht collection was published. The data compiled in Denzel (2010) show that the average annual discount in the years from 1761 to 1770 ranged from 17.76 percent (1767) to 26.13 percent (1764) with an average of 22.79 percent for the entire decade.53 Application of the discount resulted in a total expense of 165 Mark banco and 12 Schilling (line 42). In the final part of the CC, the value of the expenses was added to the total value of the transaction in Mark banco (line 44) and the bales were weighed again in local (Hamburg) pounds (line 44). From the estimated weight of 2,940 pounds, one percent Gutgewicht and 4 percent tara were deducted (line 45-48). At the very end of the CC, a price per pound was given for the 2,832 pounds of cotton that had arrived at Hamburg (line 50). The price was 14 7/16 Pfennig, resulting in a value of 1,277 Mark banco, 11 Schilling and 6 Pfennig (line 50). From this estimated value, a 13-month rebate was discounted (line 51) to arrive at a final value almost identical to the total value plus costs in Mark banco (line 53). This is the end of the CC. It indicates the costs to be taken into account and the price to be obtained in order to make a profit on the purchase. At the same time, it provides a kind of blueprint for making estimates with different quantities or exchange rates.
15To facilitate content analysis of the CCs, its main structural components were annotated and cost items were categorized. The structural components include an introductory description of the transaction with details about weighing and valuation (Table 2, A), a description of costs at origin and often a currency conversion to establish the initial value of the transaction (Table 2, B). These data were followed by a detailed listing of cost items at origin and destination, and during shipment, often interspersed with intermediate sums and ending with a calculation of the total value of the purchase (including cost items and agio) (Table 2, C). The final part of the CC provides an estimate of the preconditions for making a profit on the transaction (Table 2, D). Often, this estimate includes details about weighing and valuation at the cargo destination. In some cases, the publisher has added some additional observations about the CCs. For the categorization of cost items, commodity handling operations, such as loading or unloading, delivery to a warehouse or to a ship, packaging and preparation for long-term conservation of the cargo, etc. were distinguished from tax, insurance, money transfer and services payments, and from payments for documents or packaging materials. We have categorized these actions as “logistical services”. Furthermore, a distinction was made between costs incurred at origin, intermediate locations and destination.
16Whereas the previous section shed light on the contents of the CCs, this section digs deeper into the different transport cost items they included. Table 3 shows the total number of cost items for each CC in the selection as well as the number for Hamburg. It also specifies the costs as a percentage of the total value of the transaction. The percentages given in the table need some explanation. Calculation of the percentage shares was based on the value at origin and at arrival in Hamburg, converted into Mark banco (for example, line 5 and line 21 in Table 2). The percentage shares of the cost items were based on these values, although most of the Hamburg costs were, in fact, calculated and paid in Mark courant rather than in Mark banco. At this stage, the discounted agio (for example, line 38 in Table 2) was not computed for each of the cost items in Mark courant. Table 3 distinguishes between costs for activities directly related to the cargoes, such as loading and unloading, or delivery to the merchant’s (ware)house, expenses for documents, packaging materials, and different kinds of payments that were due before transportation, during shipment, and upon delivery of the cargo. These payments were categorized in five groups: general expenses, insurance, money transfer, taxes and payments for services. From the latter category, further details about Caplaken, Fracht (freight) and Primage are given, along with a percentage of the total freight price (Table 3, Freight+ column).
- 54 [J. A. Engelbrecht], 1772, I, p. 8.
- 55 [J. A. Engelbrecht], 1772, I, p. 8.
- 56 [J. A. Engelbrecht], 1772, I, p. 7.
17The details for the first CCs in Table 3, dealing with anchovy destined for Danzig and coming from Marseille, are remarkable in a number of ways. In the case of anchovy destined for Danzig, transport costs were the highest of all selected entries from the Engelbrecht collection, accounting for more than half of the value of the transaction. This was due to the high costs of services (22.1%), taxes (10.9%), packaging materials (11.7%) and activities related to the cargo itself (7.8%). In the CC, the purchase in Hamburg of three casks for 2 Mark and 8 Schilling each, as well as further packaging materials (emballage) for 1 Mark per cask, is specified, which explains the significant share of packaging materials in the total delivery costs.54 Interestingly, the freight costs, accounting for 9.1% of the value of the transaction, were not exceptional. Most service payments were due in Danzig for cargo delivery and storage. One further aspect of the anchovy for Danzig CC is that freight was registered as a cost item in both Danzig and Lübeck.55 In fact, in this CC, 26 different cost items were unequally divided among Hamburg (5), Lübeck (9) and Danzig (12). Coincidentally, the second highest transport costs in the selected entries also concerned anchovy. In this case, the anchovy was brought from Marseille to Hamburg and the transport costs accounted for as much as 41.4 % of the value of the transaction. Again, anchovy was transported in casks containing 12 small barrels.56 The freight cost for this cask amounted to 17.5% of the total value of the transaction, of which 1.9% were due for averie and 15.6% for Fracht (Table 3). Compared to the previous CC, taxes were much lower (1.7%), but insurance payments significantly higher (5.3%).
Table 3. Breakdown of cost items in selected CCs
Note. Own calculations based on [J. A. Engelbrecht], 1772-1774.
18Observed in isolation, these two CCs alone might give a distorted impression of the data compiled in the Engelbrecht collection. As the remainder of the table shows, the rest of the selected CCs are neither as unexpected nor as hard to explain as the CCs for anchovy, although, of course, there are significant differences between the selected commodity calculations, both in the number of cost items listed, their composition and their share in the total transaction value. This is in line with Van Tielhof and Van Zanden’s findings about freight rates. Indeed, significant differences between transport costs and freight prices are the rule rather than the exception. After all, they depended on a number of factors, not indicated by the source. These factors included, for example, differences between commodities, ports of origin or destinations and length of voyage, commodity price at the time of the transaction, seasonal differences that affected both commodity and transportation prices, different tax regimes, and the availability or not of commercial infrastructure to facilitate transactions.
19In this sense, it is worth taking a closer look at the different shares of transport costs and freight prices for cotton imported to Hamburg from different European ports (Table 3). Transport costs accounted for 9.3% of the total value of the transaction in the case of Amsterdam and 33.4% in the case of coffee imported from Livorno. This indicates that distance did play a significant role in determining transport costs. However, distance alone cannot explain the differences between the transport costs for cotton from Nantes, Bordeaux and Marseille. Whereas the total share of transport costs for cotton from Nantes was similar to that of Amsterdam (and lower than London), the costs from Bordeaux were almost the same as those from Livorno. Part of the reason is that freight costs and taxes were higher in Bordeaux than in Marseille and, especially, Nantes. The taxes paid on a load of coffee imported from Nantes represented 1.2% of the transaction value and the freight price 2.4%, compared with 4.6% and 7.3%, respectively, in Marseille and as much as 11.8% and 10.7% in Bordeaux (Table 3). Insurance rates, on the other hand, were higher in Marseille than in Nantes or Bordeaux. In the following graphs, we observe the relation between distance and insurance rates, on the one hand, and freight prices, on the other.
Figure 1. Distance in nautical miles vs. freight prices
Source. Based on selected CCs from the Engelbrecht collection.
20Even though the number of cases is low (33 CCs), four groups of CCs can nonetheless be distinguished when the distance between the port of departure (or destination) and Hamburg is plotted in relation to the share of the freight price in the total value of the transaction as found in the CC (Figure 1). The first group is the largest and the most homogeneous. It comprises CCs with a distance between ports of below 1,500 nautical miles and freight prices below 5%. Starting from the port closest to Hamburg, this group consists of Amsterdam (1.1-2.8%), Danzig (1.8%), Dover (2.9%), St. Petersburg (2.1%), London (3.4–4.5%), Le Havre (0.9%), Rouen (1.7%), Saint-Malo (3.4%), Brest (1.4%), Nantes (2.1–3.4%), La Rochelle (1.7%), Bordeaux (1.3–3.7%) and Bayonne (1.9%) (see also Table 3). The second group contains a small number of cases where the distance is less than 1,500 nautical miles but the freight price is between 5 and 15 percent. In this group, we find single CCs from some of the same ports as in the first group: Danzig (9.1%), London (7.0%), La Rochelle (6.1%) and Bordeaux (11.9%) (see also Table 3). The third group is on the right-hand side of the graph and shows a combination of long distances and freight prices between 5 and 10 percent (except 1 CC). The fourth and final group combines long distances and freight prices exceeding 15 percent of the total transaction value. The third and fourth group consist of calculations for commodities from Marseille and Livorno. Remarkably, except for one load of cotton imported from Marseille to Hamburg with a freight price accounting for 15.6% of the total transaction value, all shares of freight prices for imports from Marseille were within the same range as the imports from groups one and two, even though at least double the distance had to be covered. Whereas groups one and four correspond to an obvious situation – the longer the route, the higher the price – it is not so easy to explain the relation between groups two and three. This is due to the character of the CCs, for which the compiler gives no information about the transaction date. Since seasonality could have a decisive impact on the freight price, this is unfortunate. A second reason might be linked to the amount of cargo that a shipmaster had on board his ship. If he still had a lot of available capacity shortly before leaving the port, he might ask for a different price than if his ship was already almost full. How the freight price was negotiated for relatively small cargoes that, unlike bulk loads of grain, did not take up the entire space of a ship, is something that requires more research. Unfortunately, at this stage, it is too early to draw significant conclusions from this finding, but it seems to call for a closer look at the relation between the kind of commodity transported and the freight price.
Figure 2. Distance in nautical miles vs. insurance rate
Source. Based on selected CCs in the Engelbrecht collection.
- 57 M. A. Denzel, 2020; quote from M. A. Denzel, 2022, p. 178.
- 58 [J. A. Engelbrecht], 1772, I, p. 84.
21In addition to the relation between port-to-port distances and freight prices, it is also worth examining the relation between port-to-port distances and insurance rates. However, it should be noted that insurance rates were not always listed as costs in the CCs. In two cases, one CC for coffee exported from Hamburg to St. Petersburg and one CC for liquorice juice imported from Amsterdam, no insurance rates are given. That said, figure 2 shows a rather clear distribution of this relation based on selected CCs in the Engelbrecht collection. Three groups can be discerned. The first and smallest group is that of distances below 500 nautical miles and insurance rates below 1%. This group consists of the ports of Amsterdam and Danzig. The second and largest group contains CCs with an insurance rate of between 1.5 and 2.5%. This group includes the majority of cases and almost exclusively contains CCs with a distance between ports ranging from 500 to 1,500 nautical miles. As before, the third group is found at the top of the graph. It combines long distances from Marseille and Livorno with insurance rates between ca. 2.5 and 5.5 percent. The relation between those two parameters is more proportional than in the case of freight prices. However, it is not totally proportional. Whatever the commodities transported, it was the Mediterranean leg of the voyage that seems to have inflated insurance rates. Indeed, the commodity trade with ports in the Mediterranean Sea found itself constantly “[…] under the Sword of Damocles of the Barbary threat”.57 Unfortunately, the CCs do not provide any information about important parameters for setting the insurance rate, such as the year and season of the transaction, which complicates the comparison of Denzel’s series derived from the Hamburg price currents (Ger. Preiscouranten) from 1736 to 1859 with the CCs from the Engelbrecht collection. What the Hamburgische Waarenberechnungen do show, however, is how the sums due for insurance were established and how this results in a difference between the applied insurance rate and the insurance rates expressed as a percentage of the total transaction value. As can be seen in line 23 of Table 2, the insurance rate for the load of cotton imported from Bordeaux was 1.5%. This rate was calculated on a transaction value of 1,050 Mark banco. At the same time, line 21 of Table 2 shows that the value of the cargo was 1,010 Mark banco (Table 2). In this example, the difference between 1,010 as transaction value and 1,050 as value for application of the insurance rate is small. However, this is not always the case. In the example for capers imported from Marseille to Hamburg,58 the value for insurance was rounded up from 610 to 700 Mark banco. As a result, the insurance rate of 4.0% actually accounted for 4.6% of the total value of the transaction documented in this particular CC. It seems to have been common practice to use rounded values (1,050; 700) for calculating insurance rates, even if this often increased the amount due. Here as well, further research and the processing of more CCs from the Engelbrecht collection will shed new light on how transactions were executed and on the practical “rules of the game” for forwarding commodities between international markets.
22This paper has introduced a source that was probably well known within the Hamburg merchant community in the eighteenth century but has largely been forgotten since. Compiled by Johann Andreas Engelbrecht (1733–1803) in the early 1770s, the Hamburgische Waarenberechnungen contain a collection of commodity calculations for goods imported to or exported from Hamburg. The paper has provided some preliminary answers about how this source links up with existing research based on freight rates and has discussed the contents, structure and different cost items in the Hamburgische Waarenberechnungen. The Engelbrecht collection of commodity calculations provides a unique perspective on commercial practices by bringing together information that, if indeed it still exists, is usually scattered across different account books and the correspondence of merchants, brokers and shipmasters. In doing so, the Hamburgische Waarenberechnungen shed new light on transport costs in general and provide a more comprehensive picture of all the different minor and major cost items involved in sending goods from one place to another.
23The Engelbrecht collection provides a spectrum of CCs that is broad enough to produce qualitative statements about transport costs in Hamburg’s international trade, but is obviously too small and too limited for robust quantitative analysis of single cost items in the calculations. Nevertheless, in combination with the broad and representative scope of the source, the level of precision and detail in the CCs does provide an opportunity for gaining novel insights into the composition of transport costs in early modern trade in general, and the role of freight prices as constituent parts of these costs in particular. The results so far indicate that the source provides reliable information about freight prices and insurance rates, while adding a practical perspective to these classical topics in the history of early modern transport and trade.
24A particular strength of the source lies in its focus on the practices of transportation in eighteenth-century international trade. The CCs explicate various calculations and procedures that would normally remain “out of sight”. In doing so, the Hamburgische Waarenberechnungen offer a practical, local perspective on the “rules of the game” at European ports in the eighteenth century and provide insight into the market knowledge that merchants needed to conduct their international business. The CCs offer insight into the calculations of insurance rates, freight prices, taxes, storage costs and so on as parts of comprehensive profitability estimates. Moreover, they systematically cover diverse aspects of international commodity deliveries such as packaging, storage, loading and unloading costs, local customs, payment of shipmasters, currency conversions, weights and measures. In-depth analysis of this information beyond the preliminary insights derived from this paper is bound to deliver new structural insights into the practices of early modern (maritime) trade and transport.
25As the processing of the Hamburgische Waarenberechnungen continues, more and more data will become available that might contribute eventually to a reassessment of freight prices, insurance rates, taxes and other transportation costs in early modern international trade. It will also provide opportunities for comparing the insights obtained from the CCs about the composition of cost items at different European ports and the variety of weights, measures and currencies that merchants handled with the commercial knowledge acquired from manuals and handbooks. This, in turn, will further improve our understanding of the logistics of early modern transport and trade.