TY - JOUR
AU - Allen, Douglas W
AB - Abstract The British Navy’s Victualling Board was created in 1683 after more than a century of victual attempts rife with corruption, incompetence, and inedibles; and ended in 1832 with the creation of the modern Admiralty. Historians are unanimous that the Victualling Board was an immediate and continuous improvement, daily feeding tens of thousands across the globe. On the other hand, to the modern eye, many aspects of the Victualling Board appear odd and costly—excessive inventories, unpassed accounts, charge and discharge accounting, in-house production, mutual surveillance, and excessive punishments. Here these two observations are reconciled. I argue that the Board’s organization was a clever response to pre-modern transaction cost problems; that is, the Board was designed to create compatible incentives subject to the measurement constraints of the time. It [victualling] was a problem that involved not only shipping, but also strategy and communications, and in struggling with it the administrators groped in the dark: they worked from inventory reports that were months out of date; they could not calculate the amount of spoilage; they could not be sure that the latest shipment was not lost to the enemy or the elements. But they followed a method which, in view of the uncertainties, made sense and generally proved successful. [Baugh 1965, p. 444 (emphasis added)] 1. Introduction Most layman perceptions of food and drink aboard an eighteenth century British warship are based on pop-culture references with Victorian roots. John Masefield’s Sea Life in Nelson’s Time (1905), for example, described the food as “always bad, and sometimes villainous” with weevil laced biscuits eaten “at night, when the eye saw not, and the tender heart was spared.”1 But such sentiments turn out to be false. Over the past fifty years naval historians have shown that food provisions were supplied in sufficient quantities and qualities at the appropriate times in order to “sustain the fleet” consistently for the last 150 years of the age of sail.2 Given the amazing success of the British Navy over this period, victualling must have been adequate.3 It certainly would have been impossible for the British Navy to have developed their Western Squadron in the mid-eighteenth century which was able to stay at sea for months at a time, if victual supplies had been insufficient.4 But if sufficient, then the British Navy’s daily feeding of tens of thousands of souls while “groping in the dark” seems like a miracle of Biblical proportions.5 A feat all the more impressive given that virtually no British naval tactical operation of the latter eighteenth century was lost for want of provisions.6 Indeed, from the creation of the Victualling Board in 1683 until its elimination 149 years later in 1832, there was an immediate reduction of costs and continual improvement in the supply of provisions to the point where they “had been brought to a degree of perfection never known before.”7 At a time of difficult and often too costly measurement of fundamental quantities such as time, location, volumes, temperatures, and distance; in an era of unreliable and highly variable communication, embryonic methods of accounting, limited and indirect parliamentary oversight, ships propelled by the vagaries of wind and weather; and within the social context of “old corruption” which ruled the roost over talent and merit, such an outcome seems all the more impressive.8 Here I provide an explanation for why the Victualling Board was an improvement over earlier attempts, why it was able to operate so well for so long, and why their methods eventually came under attack and were replaced at the end of the pre-modern era. In short, until the end of the eighteenth century, the Victualling Board used simple organizational methods that effectively dealt with the measurement problems of the pre-modern world without bankrupting the treasury. The many nuanced features of the Board were the result of this process, and although the “back of the envelope” methods make little sense in the modern world that began to emerge at the end of the eighteenth century, they were necessary at the time because modern methods of measurement were unavailable.9 By 1800 many pre-modern transaction cost problems had disappeared as a steady stream of measurement improvements in the areas of clerical and manual work, scheduling, record keeping, and planning developed.10 These made the earlier organizational solutions for managing these problems obsolete, and led to radical changes in the organization of the Victualling Board towards modern offices of wage-only employees.11 This theory is tested in several ways. First, it is documented that the rational objective of “abundant stores” was met. Second, the institutional aspects of the board—the use of patronage, venality, and the “rule of three”—is argued rational in light of the measurement problems of the time. Third, the removal of venal office clerks is shown to correspond with office measurement innovations. Fourth, the choice between supply contracts and in-house production is shown to depend on the measurement costs of quality across the four main products: beer, beef, biscuits, and water. Finally, evidence from the average cost of victualling over the Napoleonic Wars shows evidence of better measurement at the Victualling Board. This approach—that historical institutions are “the solution to some problem that is preventing people from achieving higher production and consumption” (Ogilvie 2007, p. 651)—is common among economists. The modern “endogenous” treatment of historical institutions goes back, at least, to North and Thomas (1971), and over the past forty years everything from serfdom and guilds, to dueling and usury have come under analysis.12 Early work by North and others claimed or implied that all institutional outcomes were first-best efficient, in that they were designed or evolved to maximize wealth.13 Few today would hold such a position. Rather, most claim that the institutional features in question (the humanly created, immaterial, set of economic property rights) are the result of some type of maximization process subject to some set of constraints. Depending on what gets considered a constraint determines whether any given set of institutional details are considered “constrained efficient” or not. Hence, Ogilvie (2007) considers “accidental events and personalities,” “cultural beliefs and values,” and “conflicts over the distribution of resources” as three alternative theories of institutions in contrast to wealth maximization.14 Others, following more closely to a Coasean logic, see institutions as the result of wealth maximization subject to all relevant constraints—which might include accidents (nature), beliefs, and fights over the gains.15 Here, the latter approach is taken, and no value judgment with respect to the institutional arrangements of the Victualling Board is made. Rather, the Victualling Board solutions are considered rational and evolved under the real world circumstances of the time: the Victualling Board did the best they could under the circumstances they faced. Among many elements, these circumstances included the violent competition of other navies, the cultural constraints of the time, the massive role of nature, Parliamentary and Navy oversight, and the general institutional context of the British “public service.” Given the size of the navy and the importance of victualling in its operation, such a study provides an important example of understanding the role of how these sets of complicated constraints can manifest in institutional outcomes. Throughout the age of fighting sail, mitigating transaction costs that arose because of the conditions of the time, was a first-order goal of the British Navy.16 Here I show that concern over these costs was not limited to the incentives onboard ships, but also extended to victualling logistics. As such, this study extends the economic analysis of institutions in history.17 2. Victualling Board Historical Essentials A navy, to paraphrase Napoleon, “sails on its stomach,” and as the English pre-modern navy began to develop in the early sixteenth century, victualling was of vital importance and organized directly through the Crown.18 In 1546 Henry VIII established the Navy Board to run all aspects of the fleet, and by 1550 this included a department of victualling to deal with private contractors.19 For the next 137 years feeding the fleet was a hodgepodge of various ad hoc combinations of contractors, special boards, and admiralty interventions, often at arm’s length from the Navy Board and seldom with any institutional memory—none of which worked well together.20 A fundamental problem existed around the governance of victual supply. In this period, the Navy Board relied on contracts with a small number of major contractors who were paid fixed amounts per man, per day. Given the difficulty in measuring the delivered quality and quantities accurately, this created an incentive to chisel on both dimensions. The problem was exacerbated by the fact that most of the contractors held the naval victualling offices entering into the contracts. Hence the incentives of those in the naval office were in conflict with the interests of the Crown, and led to habitual under provision and over charging.21 Incentives were also misaligned at lower levels of the victualling chain. For example, pursers were paid in proportion to the number of men on the ship’s books, which encouraged false mustering, often in collusion with the captain.22 Pursers also purchased stores with no checks on their behavior or the quality of purchase, and over-payment resulted.23 Finally, given the Crown “paid its servants and suppliers partially, tardily, seldom, or not at all,” contract markets were under-developed, making merchants reluctant to extend themselves as might be required.24 Victualling fraud was rampant among contractors, captains, and pursers because the early system “not only encouraged… fraud but virtually enforced it.”25 As a result, prior to the creation of the Victualling Board, the system was highly defective with victual failures common: squadrons were delayed, ships forced off station, and operations and blockades terminated early.26 The idea of an extended cruise was not a possibility, and ships had difficulty operating in the North Sea for more than a few months, despite the proximity to home ports.27 In addition to restricted tactics and fraud, food quality was lacking with weevil laced bread, fouled fish, rancid meat, sour beer, and maggots commonplace. Reporting at the end of this pre-Victualling Board era, a captain noted: “… in several of the butts of beer, great heaps of stuff was found at the bottom of the butts not unlike to men’s gut’s…”.28 By the mid to late seventeenth century the system was in a “state of collapse.”29 At the restoration in 1660 the Navy Board set up an internal victualling board, but despite some improvements problems in successfully delivering a given amount of food, of a minimal quality, at a specific time remained. Finally, in 1683 an organizational relationship between the Navy Board and the Victualling Board was reached that would remain in place until 1832 when the Admiralty took direct, modern, bureaucratic control over victualling.30 Under the new system a separate Victualling Board was established directly under the political Admiralty Board, headed by the First Lord of the Navy, but nominally and in practice under the Navy Board. The fundamental objective of the Victualling Board was to supply food and drink in a timely manner, and this was accomplished by (1) collecting wholesale supplies and issuing them from their own stores (including some processing), (2) contracting with hundreds of private individuals or syndicates who operated their own stores, or when necessary (3) allowing pursers/captains on ships to purchase in foreign ports. The collecting of wholesale supplies was done either through private contracting or through the use of agents working on commission.31 The Victualling Board paid contractors mostly in interest bearing negotiable victualling bills, that were generally accepted around the world by the late eighteenth century. This was made possible by Parliament’s take over of state finances in the latter half of the seventeenth century. All loans were registered in an unalterable Treasury ledger where they were dated and numbered sequentially, and then paid back in order.32 This system was adopted by the navy in the 1690s and provided market confidence in their debt instruments, allowing the navy to contract for provisions at much lower costs.33 It is most notable that the technology of food preparation, preservation, and storage was unchanged from the beginning of the English Navy (c. 1500) until the end of the boards in 1832.34 Systematic changes in the provision of food would come with the introduction of canning, refrigeration, and steam propulsion, but these changes took place later in the nineteenth century. Also, the public/private partnership between the navy and contractors in the supply of victuals existed from the beginning of the Tudor navy, and remained when the Admiralty took over in 1832.35 What changed in 1683 was not a matter of technology or complete vertical integration. Rather, the improvement in production after 1683 was due to matters of internal organization that encouraged victuals to arrive on time in sufficient quantity and quality. The organizational cornerstone of the new Victualling Board was the division between the commissioners and the clerks. The Victualling Board was headed by seven salaried commissioners, who throughout the life of the board were chosen by the Admiralty through patronage.36 The office was a matter of private interest, and remained theirs until death, they chose to leave, or were found in some type of misconduct.37 Although a commissioner of the Victualling Board was not a great office, it was still sought after among the lower gentry and officer corp because income was earned through the sale of lower offices.38 Each commissioner was responsible for one of the major operations of the board: cutting house, brewhouse, cooperage, mills, and so on, and together they met to allocate contracts to private individuals. The Victualling Board also contained clerks—first Clerk of the Paymaster, First Clerk of the Pay Branch, Second Clerk, etc.—who purchased their position and held them as a private property.39 After Parliament had made its “estimate,” funds were transferred to the Navy Board, and then subsequently transferred down to the various branches, including the Victualling Board. Clerks were responsible for maintaining the accounts, which used a “charge and discharge” system that allowed the clerks to keep track of when any transaction was completed.40 Although clerks received a nominal salary, much of their income came from fees and gratuities that were paid out of the funds granted as accounts were closed. Furthermore, until the accounts were closed and all fund disbursed, clerks could hold funds in their own private accounts and earn interest.41 The role of contracting with independent private parties can hardly be overstated; indeed, Knight and Wilcox coin the term “the contractor state” to describe naval procurement in the eighteenth century. In 1800 victualling accounted for 35% of naval expenses, and yet the Victualling Board’s central office only had 105 employees in 1805.42 Thus, although the Victualling Board processed and purchased some provisions directly, their major function was to administer, monitor, and assemble products supplied by others.43 The bulk of victualling in-house processing took place at the Deptford yard near London, although Portsmouth and Plymouth were also major yards. 3. Victual Production: Timing, Measurement, and Uncertainty Understanding why the early system failed, and why the Victualling Board succeeded until the modern era, requires an understanding of the critical production constraints in the pre-modern era. Aside from the size and scope of the navy’s victual challenge, three features made it organizationally difficult. First, the urgency of adequate victual arrival was asymmetric around the optimal arrival date. Second, over the course of the long eighteenth century, the difficulty in directly measuring much of the victualling process, meant an inability to monitor in any modern fashion. Finally, the role of Nature in production was enormous. These three factors constrained the organization options available to the Admiralty. 3.1 Timing of Arrival With naval victualling the cost of failure could be extreme. Ships with consistently late and/or poor quality provisions could end up with diseased and decimated crews, and a potential mutiny across the entire navy.44 Strategically, a late delivery of food or water could mean the loss of a battle, a blockade, or perhaps an empire; and in times of war, the cost of a victualling failure would be greater than in peacetime. Such production payoffs, in an extreme, could be represented by figure 1, where there is an optimal arrival date a*, and the victual payoff is asymmetric around this date. An early arrival has a lower payoff because it must be stored, but a late arrival has a payoff of zero. Figure 1. View largeDownload slide Asymmetric timing Figure 1. View largeDownload slide Asymmetric timing The probability of provisions arriving to a ship by a given date is given by a density which depends on efforts made by the Victualling Board. An increase in effort increased the probability of provisions arriving in time. Since the Victualling Board had a good knowledge over when provisions had to arrive depending on the context, the value of a* was generally known. In deciding the timing of victual arrival, the Victualling Board had to consider the asymmetric consequences. 3.2 Measurement in the Pre-modern Era The second critical feature of victualling production—common to much of life in the pre-modern era—was a difficulty to accurately measure fundamental quantities such as time, distance, volume, and weight.45 Such an inability created a serious constraint on what type of organization was even possible for the delivery of provisions. For example, consider the difficulty of accurate time measurement. Although a horological revolution took place between 1650 and 1750 that saw errors in clock-time fall from 500 s per day to about 0.1 s per day, the precision of clock-time only slowly became culturally embedded.46 For ordinary folks of the early eighteenth century, their day was regulated by sunrise, noon, and sunset and the town bells that marked those moments. Clocks existed, but they were large communal, mostly aural, affairs that were not accurate to a practical degree for the Navy. Watches, even with a second hand, were not uncommon, but not readily accepted by the average Englishman. For the Victualling Board, the problem of time measurement was of primary consequence. First, even though the first chronometers (an extremely accurate clock) were available in the 1760s, by 1802 only 7% of naval ships had one.47 The absence of a chronometer required longitude at sea be estimated using a sextant and lunar distances, or dead reckoning—both methods subject to large error depending on the conditions. This meant naval ships often followed a coastline until they reached a known latitude before venturing into open waters—not always with successful outcomes. Second, poor longitude measures resulted in poor maps, and so ships often were lost and delayed, run aground on unmarked shoals, and unable to readily rendezvous with other ships in open seas. These problems exacerbated the Victualling Board’s ability to accurately schedule the arrival of a ship at a distant harbor, or accurately predict when (or if) ledgers would arrive to payoff a ship. Accurate clocks were also necessary for the standardization of measures such as distance, weight, and volumes.48 Hence, until the mid-eighteenth century when pendulums and balanced springs reduced the variance in time keeping significantly, and until the late eighteenth century when accurate clock-time had penetrated most parts of society, the concept of being at a specific place at a specific time was not part of the general culture. Nor was the ability to consistently measure distances, weights, and volumes. Life was, as it had been, a matter of “by guess and by God.” Such an environment affected the Victualling Board’s everyday operations. Everything—the collection of food inputs, the measurement of grain piled in storage, the ability to maintain consistent temperatures in food production, the timing of delivery at sea and on land, the knowledge of travel conditions, the monitoring of the arrival and departure of clerks, and fine-tuned quality measurement—was done in an environment where modern abilities to measure basic fundamentals, and therefore directly measure performance, were weak. In contrast, the end of the long eighteenth century saw radical changes in the ability to measure fundamental quantities. The chronometer, barometer, mercury thermometer, coke blast furnace, and reflecting telescope are a few of the measurement devices that were perfected by the end of the century. Along with theoretical and scientific developments, these innovations allowed for inexpensive and standardized printing, machine tools, and processes. These changes allowed the ability to connect inputs with outputs in a reasonable way for many occupations, and allowed the notion of accountability to become embedded in the culture to a degree previously unknown. As a result, there was an increase in official records, a desire for sound information, an emphasis on statistical measurement in administration, an understanding of being at a specific place at a specific time (as in a modern labor contract), that had never existed before.49 3.3 Nature Everywhere The final element of victual production, again common to much of pre-modern life, was the immense role of Nature. First, Nature constrained victual production through seasonality. Victualling was fundamentally about food, and therefore, was subject to seasonal patterns. For example, pork and beef could only be salted in cool weather, and so had to be done in the fall or winter.50 Second, Nature was often used as a factor of production. In the pre-modern era mechanical power was mostly absent, and so Nature was the major supplier: animals, wind, and water were used for transport and the movement of ships and mills; and the sun was used for such things as drying grain. Finally, Nature provided random shocks to output because it was usually infeasible to bring production indoors, inputs and outputs were often exposed to the elements. The critical economic element of Nature is that it is highly variable. Crops are seasonal, but starting and ending dates vary year-to-year. The sun might be used for drying, but weather variations lead to variations in the timing and quality of the output. Hauling, transport, and other movements of cargo on land are done by animals and humans, which vary in ability. Ships propelled by wind and tide could get stuck in base or the channel for weeks, and, of course, many ships never arrived at their targets.51 4. The Rational Victualling Strategy The potential catastrophic cost of being late determined the objective of the navy with respect to victuals: minimize the probability of late victual arrival, subject to the measurement constraints. Since it was impossible to finely tune reliable delivery times, this could only be accomplished by making certain that stores were available on hand. Such a “just-in-case” production strategy required strong actions to keep stores inventoried to avoid a victualling failure.52 Achieving such an objective, however, was constrained by the measurement environment. Since the pre-modern era was characterized by large roles of Nature and a general inability to remove or measure its effects, there were opportunities for all forms of malfeasance on a large scale.53 In naval victualling operations, this meant the potential for various types of shirking, theft, and dereliction of duty was omnipresent at every level of the organization. Without the appropriate organizational design to manage incentives a victualling failure was likely, and a failure to “arrive-in-time” and in sufficient quantity and quality, could mean a disaster for the navy and Crown. Prior to the creation of the Victualling Board, the English Navy was unable to manage these transaction cost problems, and as a result it was generally unable to achieve the objective of timely arrival. The 1683 Victualling Board solved the problem in three ways. First, like other pre-modern institutions of the time, the Victualling Board was organized by offices that were either held by patronage or venality. Venal offices worked well when their owners had incentives that aligned with those of the Crown, and the profit residual effectively policed behavior. Patronage appointments were made in cases where the private incentives of the office holder were at odds with the Crown. Here aristocrats—individuals who made enormous investments in sunk physical and social capital—were given offices and operated under the threat of great losses if caught in any type of malfeasance. In both cases, often no forms of modern monitoring took place. Second, within this patronage and venal structure, the Victualling Board focused on those specific margins that could be measured, and designed a victualling institution around those margins to ensure that stores were always available “justin-case” nature dealt a large negative shock. Finally, in dealing with contractors and production, the Victualling Board used incentive compatible contracts or in-house production, depending on its ability to measure the quality of the good being procured. Despite being an age of difficult measurement, there were three output measures that were observable in a meaningful way. First, it could be verified whether or not a shipment arrived on a specified date (either at a yard, dock, or station).54 Why a ship arrived on a particular date (whether early or late), could not be verified, but the actual date of arrival could be known, and compared to the expected date (a*). Second, when bundled and packaged a specific quantity of provisions could often be counted and this count could be known. The actual quantity of flour in casks, biscuits in bags, or wine in butts had variance, but the number of casks, bags, and butts could be verified. And finally, it was possible to verify that the food and water delivered met a minimal quality threshold: fit for human consumption. If seamen could stomach the food, the minimal quality threshold was met.55 Hence, it is predicted that a rational Victualling Board would operate such that ships were never in dire want of supplies. That is, provisions should have been made in an early abundance to meet unexpected demands throughout the year, and supplies should have been inventoried at the various stations around the world to meet virtually all contingencies. Furthermore, avenues should have been available to satisfy the demands of ships when the Victualling Board could not meet demand. At the same time, to the extent they were able, the Victualling Board should have mitigated transaction costs at every stage of production by monitoring the three margins available, and designing contracts in ways that mitigated measurement problems.56 5. Evidence of Abundant Stores A rational victualling operation would have created a large inventory in order to distribute supplies, even in cases where demand suddenly increased. Although this objective was not reliably achieved prior to 1683, the Victualling Board was able to accomplish it. Interestingly, a policy of excessive inventory was in conflict with the official planning policy of the Navy which had carried over from Tudor times. Since the beginning of the age of fighting sail an instrument called the “Declaration of Victuals” had been issued once a year (usually in September) by the Crown estimating the number of men the navy was required to feed for the following year. Prior to the establishment of the Victualling Board, the Declaration was a binding constraint on the Navy. Unfortunately, and not surprisingly, during the sixteenth and seventeenth centuries, the Declaration was often inaccurate.57 The Declaration also tended to be low, and this resulted in victualling shortages which led to poorly fed seamen and victual failures that “repeatedly frustrated and constricted the operations of the English fleet.”58 The declaration was “something of a farce.”59 Once the Victualling Board was created, the Declaration of Victuals continued pro forma, seldom changed from year-to-year, and was ignored by the Admiralty. Instead, the Admiralty issued its own instructions for excessive supplies. Thus, when the 1738 declaration was not issued until March of 1739 (when it would have been of no use) and only ordered food for 14,000 men, the Victualling Board was still able to feed 24,000 men that summer when England fought with Spain in the War of Jenkins’ Ear.60 Given that beef and pork had to be salted during cool months, a critical aspect of the abundant stores practice was that “they already had a respectable quantity of provisions on hand” because “… superabundance was the only insurance against the uncertainties.”61 By the mid-eighteenth century, “a general order was given which allowed captains in home waters to take on three months’ provisions, as required, without any order from the admiralty…. The Admiralty was determined that there should be no delays for want of proper pieces of paper.”62 The practice of excessive supplies was common at the Victualling Board. As mentioned, the Western Squadron was constantly at sea, and required continual victualling. The directions to the contractors stated that they were to “be prepared for all eventualities.”63 Foreign yards were to keep large inventories, shipments and deliveries abroad were to be frequent and smaller to avoid a disaster if one large ship failed to arrive, and stores were always to be “quantitatively sufficient and qualitatively sound.”64 The result was that “Food shortages… common during the wars of the League of Augsburg and Spanish Succession… became… altogether exceptional during the Seven Years War”.65 The focus on abundant stores for all contingencies explains a financial puzzle at the Victualling Board. Until the end of the eighteenth century it had access to almost unlimited and unapproved funds; that is, the Victualling Board did not require explicit Parliamentary approval for debts incurred.66 As noted, at the end of the seventeenth century the Victualling Board began to issue bills of exchange that were discounted and traded as currency. There was no accounting, periodic checking, or parliamentary approval required. As long as there was no victual failure, “parliamentary unconcern was exhibited.”67 Nor was there regular parliamentary auditing of issued funds held by the Victualling Board. The result of this was that the Victualling Board always had funds on hand to meet unexpected demands for stores. For example, the 1781 commission on public accounts found that the Board “… guard[s] against the hazard of a demand upon an exhausted fund” by keeping “a very large sum… constantly on hand”. Furthermore, until an account was closed, funds allocated by treasury were retained by the clerk or other officer, and could be used for any operation.68 For the first one-hundred years of operation, as long as the “front-end” of operations were satisfied, no concern was given to the loose accounting. If judged by today’s standard, these practices appear as misuse of public funds.69 However, such a conclusion is not warranted. The funds had to be used for victualling, and theoretically, all accounts were eventually closed and “accounted” for. The system allowed the Victualling Board to be financially unconstrained by a fixed budget over the short-run, and therefore it was always able to accommodate changing demands. 6. Evidence of Mitigating Measurement Costs The objective of abundant stores was the result of the pre-modern environment that made modern “just in time” production impossible. However, as argued, this goal was achieved within an organization that was also designed to mitigate the measurement costs of the pre-modern world. This explains why commissioners were patronage appointments, why clerks held venal offices and why these offices were removed once measurement costs were reduced, why the victualling board used a “rule of three,” and why some goods were produced in-house while many were supplied by contractors. 6.1 Patronage Among Commissioners As explained in Allen (2005, 2012), there were two institutional solutions to the problem of “public servant” measurement in the pre-modern era, and both of these were established with the creation of the Victualling Board in 1683. The first was a system of patronage that relied on “watchdogs” and “mutual surveillance,” combined with enormous penalties if caught for misconduct. Patronage was a system whereby trusted individuals were given an office, and expected to act in the interest of the patron (ultimately the Crown). Trust resulted because these individuals were required to post enormous performance bonds in the form of sunk social and physical capital.70 As long as the office holder performed, they earned a return on their sunk investment, but if caught not performing they were outcast and the investment lost.71 Patronage was used when the incentives of an office holder who might otherwise be a residual claimant, were in conflict with the objectives of the Crown. For the victualling commissioners, the position of an administrator overseeing the buying and selling of stores with Crown money, clearly was a case of incompatible incentives. Had the commissioners earned their keep by reaping the difference in what was paid out to suppliers and charged to the navy, there would have been ample opportunity for fraud in the form of exaggerated costs, and an under provision for the crews. Indeed, such a potential still existed with a Captain’s purchase of stores in foreign ports outside of naval stations, which was a constant source of corruption. In fact, the largest number of convictions for fraud in the eighteenth century was between Captains colluding with their pursers.72 This had also been the central problem prior to the establishment of the Victualling Board when naval office holders were also victual contractors, and when ship’s pursers were given funds based on the ship roster. The 1683 Victualling Board solution created commissioners who were not contractors, on a fixed salary, and granted the office through patronage. The commissioners essentially owned the offices beneath them, and these offices were sold to clerks who paid a “premium” to the commissioner.73 Although their nominal salaries were often lower than the income received by the clerks through fees, the fees were capitalized into the premium paid for the office. Hence, the value of a patronage office depended on the overall operation of the organization, and this provided the commissioners with an incentive to run the Victualling Board in a way that maximized the value of the offices beneath them.74 Commissioners were given guidelines on performance, and could be removed from office if caught not acting in the interests of the Crown.75 Rather than an example of “old corruption,” the commissioner position was designed to provide trustworthy performance in a situation where residual incentives were incompatible and direct monitoring infeasible. 6.2 Venal Clerks and Minimal Accounting The second pre-modern method of dealing with “public servant” measurement issues was venality. Venality was a system of private sale of offices to individuals, who would then provide service for fees and manage their office as a private property. Venality worked well when private incentives to maximize the wealth of the office holder aligned with the objectives of the Crown. Such was the case with the clerkships created with the Victualling Board. Until the late eighteenth century, virtually all of the offices beneath the commissioners were private holdings. The clerks made their incomes through salaries (often nominal), fees, emoluments, gratuities, and perquisites—many of which were paid out after the accounts were passed. Prior to the creation of the Victualling Board in 1683 clerks were often treated as “employees,” which contributed to victualling failures because of an inability to monitor. The venality introduced in 1683 allowed clerks to own and manage their offices as a private monetary interest, providing an incentive to ensure that the accounts were eventually closed. Because the incentives of the clerks were aligned with the Admiralty—they were paid when the accounts were closed—they had a strong incentive to complete their task, and did not require outside monitoring. Closing (or passing) an account simply meant that several clerks had verified the contracted quantity of victuals had been delivered in time, and found fit. This was done by receiving the ledgers (accounts) of ship pursers, stating that some quantity of victuals had arrived at some time and in an acceptable quality. When all of the ledgers could be cross checked, the account would be closed, and final payments made.76 In the event of being caught in some impropriety, punishments in the form of lost offices, lost political influence, and criminal sanctions could be imposed.77 The fact that the clerks were “self-monitored” by a residual explains why they used a very minimal type of accounting system. For the first one-hundred plus years of the Board, clerks used “charge and discharge” bookkeeping which had no element of auditing or control present. Indeed, auditing was impossible until the account was closed.78 However, auditing was unnecessary because the goal of the clerk was to close the account as soon as possible in order to be fully paid, which was the same goal of the Crown—the incentives were aligned. Likewise, since the residual claimant clerks required no audit or inspection, the fact that their bookkeeping system did not easily allow for periodic third party monitoring was irrelevant. Such as system was not perfect, but the elimination of practices that might be considered fraudulent in the modern world, was never within the grasp of the Victualling Board nor any other civil department at that time.79 Thus the Victualling Board was not a modern bureaucracy, but rather a pre-modern institution designed with the conditions of the day in mind. Patronage commissioners and venal clerks, each had incentives to operate the Board in a way to achieve the objective of abundant stores. 6.3 The “Rule of Three” Although the Victualling Board “groped in the dark” they were not totally blind and without reference when it came to measurement of some things. Indeed, the board used a number of methods that might be described as “crude” measurements, or “back of the envelope” proxies. A common institutional feature found throughout the English civil administration during the pre-modern era might be called “the rule of three,” which was a system of mutual surveillance and reporting.80 When the Crown was interested in the truthful revelation of information it often placed three men, each with different career incentives, in charge of recording the information. Given the different career incentives, collusion at the expense of the Crown was difficult because what benefitted one often hurt another, and therefore the Crown trusted the information. For example, the three men on the quarter-deck of almost every naval vessel included a captain, master, and first lieutenant. Through their independent logs, and by order, each kept detailed records of every aspect of the ship, and thus independently monitored each other. At the end of a commission, each separate journal for the ship was returned to the Admiralty for inspection. Inconsistencies in the journals led to investigations and punishments.81 Throughout the operations of the Victualling Board the “rule of three” was used often. As noted above, the bulk of provisions were acquired through contracts with private individuals, and agents for the navy commissioner’s were in a position to collude with contractors at the expense of the navy. However, all formal business of the Victualling Board, including signing orders and making contracts required three commissioners in attendance.82 Within the victualling office itself clerks were divided into three groups: receiving, issuing, and expending.83 These clerks kept three records of the same contract until reforms were brought in during the Napoleonic wars.84 Although this duplication has been criticized as inefficient, having duplicate records was a type of “rule of three” check on the dealings of the various clerks who operated quasi-independently within their offices. When a purser purchased from local farmers when away from a station, he had to be accompanied by two officers from the fleet who would certify the price paid. These transactions were thus entered into three separate ledgers and ultimately cross checked at the Victualling Board before payment was made. According to Macdonald, this system of purchase “did not cause any acute problems.”85 There were separate purser books kept of ship-to-ship exchanges, and three men (the purser, captain, and master) had to verify the amounts purchased.86 Finally, if victuals were to be condemned for low quality a captain could order a survey be performed by three officers: often masters, drawn from three separate ships, and none from the ship where the victuals spoiled. Nothing was to be thrown overboard, if possible, and the provisions were to be returned. The Victualling Board would investigate and make the purser, navy, or contractor pay. Condemnations seldom were above 2%.87 No doubt collusion was not eliminated, but the simple act of having three different individuals in charge of critical information, when all three had different career incentives, allowed the Victualling Board to trust the information it received. 6.4 Contract Choice and In-House Production As noted, the Victualling Board entered contracts with hundreds of suppliers to victual the fleet. Knight and Wilcox (2010) separate the contracts into two major forms: “supply contracts” and “agency contracts.” When goods were not purchased on the market, they were produced in-house, and the Victualling Board always had yards where animals were slaughtered, biscuits baked, and beer brewed. Macdonald notes the obvious here: “… one might wonder why the Victualling Board continued to pack their own meat when it was so freely available from Ireland at a reasonable price.”88 Once again, mitigation of costly measurement explains the Victualling Board’s choice of contract type and in-house production. For supply contracts, advertisements would be placed in newspapers and other outlets to seek tenders, sealed bids would arrive within three weeks and put into a locked box at the victualling office to be opened on a specific day in front of the bidders. If the Victualling Board decided to go with contractors they chose the lowest bid subject to an assessment that the contractor could carry out the arrangement. Contractors would often post a surety for performance and swear an oath regarding quality.89 Commissioned agents, on the other hand, acted as quasi-employees for the Victualling Board and purchased on behalf of the board on the open market. Unlike the competitive contractors, who tended to be small merchants with no permanent connection to the navy, agents were (according the Victualling Board) “… persons of reputation, integrity and extensive commercial connections…”.90 Like the commissioners, agents were individuals who could be trusted because they stood to lose considerable wealth if caught in malfeasance. If the Victualling Board behaved rationally with respect to measurement costs, then a number of refutable implications arise with respect to contracting and production. First, the Victualling Board should have produced goods in-house when it was unable to measure the quality of those goods when produced in the market. Second, if production in-house was infeasible and market quality measurement costs still high, trusted agents (market specialists) should have been used to purchase the goods in the market. Third, competitive spot contracts should have been used in cases where quality measurement costs were low. Finally, if measurement costs were falling throughout the long eighteenth century, it would be expected that the use of trusted agents and in-house production should both fall over time, and be replaced by simple contacts.91 Here I use the representative sample of contracts from 1789 to 1818, collected by Knight and Wilcox from navy contract ledgers to test these implications.92 Although there are sixty-five products contracted for in the data, the vast majority of contracts are for a few products. Here, I concentrate on the four major components of the naval diet: beer, oxen, biscuit, and water. Beer quality was a difficult to measure. Brewing was an ancient art practiced to produce daily rations of beer at a time when water was often not safe to drink. High quality beer that would not go sour over time could be produced; however, in the long eighteenth century there was no ability to test for shelf-life in the finished product.93 Thus, market beer brewed in a manner to keep on a long voyage was not verifiable to the Victualling Board, creating a classic “lemons” problem. As result, beer should have been produced in-house at the victualling yards. Cattle were purchased in the market and brought to the yards for processing because it was infeasible for the Victualling Board to raise cattle. Deptford, for example, had a slaughter house that could hold 260 oxen and 650 hogs at one time. Interestingly, the Victualling Board required the cattle to be delivered as live oxen, which are steers (i.e., castrated bulls). The Victualling Board also required the animals to be “fat… good and fit.”94 Low quality meat, like beer, spoils easily on long voyages. Low quality meat, however, can be difficult to measure when the animal is on the hoof. Oxen solve part of the problem. An ox grows to full size within less than two years, and of course produces no milk or offspring. Hence, no farmer keeps an ox for more than the time it takes to grow to its optimum size, which means an ox is always young beef. On the other hand, cows and bulls can be kept for up to ten or a dozen years, at which point they can be slaughtered to produce low quality old beef. An opportunistic beef supplier delivering beef already butchered, could purchase old cows and bulls under the pretense of young oxen—once dressed it would have been almost impossible to determine the cause of the low quality meat at the time of purchase. However, when live, an ox is easy to identify, and so an ox increases the likelihood of high quality beef.95 Still, oxen could still be unhealthy, poorly fed, and not fit for service in the navy. Animal health was likely a high cost measurement case, and so the Victualling Board would have required more assurance when contracting. Hence, it is predicted that commissioned agents should have been used to acquire oxen. These agents would have been specialists in the cattle market, but they were also persons with substantial social capital and sunk commercial investments in reputation. As a result they could be trusted by the Victualling Board. The actual production of salted meat also had a serious measurement issue. Salting had to be done during the cool months of the year in order to produce quality meat, but once salted the date of production was not known. Thus, the Victualling Board should have engaged in in-house production of salted meat, or used commissioned agents to purchase it.96 Finally, biscuit and water were two simple products that have relatively low measurement costs in determining their quality. The Victualling Board had a number of simple tests of biscuit quality, and Knight and Wilcox note several cases where the board was able to identify low quality products before they were sent to sea.97 Furthermore, the victualling yards always produced some of their own biscuit which was used to benchmark quality.98 Hence it is predicted that these two goods should have been supplied by contracts compared to beer or beef. Table 1 provides the definitions and means of the variables used to test these predictions. It shows that biscuit and food ingredients were the most common items contracted for in terms of the number of contracts, while oxen, beer, and water accounted for very few contracts. Table 2 provides some multinomial logit regression results, where the dependent variable CONTRACT takes on the value zero for in-house production, one for supply contracts, and two for agency contracts. The reference category is supply contracts.99 Therefore, columns (1) and (2) show the results for in-house production relative to supply contracts, and columns (3) and (4) show the results of commission agents relative to supply contracts. The difference between regressions (1)/(2) and (3)/(4) is simply a matter of controls to show robustness. Table 1. Definitions of variables Variable name  Definition  Mean  CONTRACT  = 0 if produced in-house, 1 if supply contract  0.782    = 2 if commission agent.    BISCUIT  = 1 if contract was for biscuit.  0.307  OXEN  = 1 if contract was for oxen.  0.009  BEER  = 1 if contract was for beer.  0.017  WATER  = 1 if contract was for water.  0.003  YEAR  = 1,789 to 1,818  1,805  SALT MEAT  = 1 if contract was for salted beef, pork, or suet.  0.061  NOT FOOD  = 1 if contract was for non-food items, e.g., casks, hoops, etc.  0.022  FRESH  = 1 if contract was for fresh meat, bread, or vegetables.  0.127  INGREDIENTS  = 1 if contract was for flour, pease, oatmeal, vinegar, wheat.  0.282  Variable name  Definition  Mean  CONTRACT  = 0 if produced in-house, 1 if supply contract  0.782    = 2 if commission agent.    BISCUIT  = 1 if contract was for biscuit.  0.307  OXEN  = 1 if contract was for oxen.  0.009  BEER  = 1 if contract was for beer.  0.017  WATER  = 1 if contract was for water.  0.003  YEAR  = 1,789 to 1,818  1,805  SALT MEAT  = 1 if contract was for salted beef, pork, or suet.  0.061  NOT FOOD  = 1 if contract was for non-food items, e.g., casks, hoops, etc.  0.022  FRESH  = 1 if contract was for fresh meat, bread, or vegetables.  0.127  INGREDIENTS  = 1 if contract was for flour, pease, oatmeal, vinegar, wheat.  0.282  Table 1. Definitions of variables Variable name  Definition  Mean  CONTRACT  = 0 if produced in-house, 1 if supply contract  0.782    = 2 if commission agent.    BISCUIT  = 1 if contract was for biscuit.  0.307  OXEN  = 1 if contract was for oxen.  0.009  BEER  = 1 if contract was for beer.  0.017  WATER  = 1 if contract was for water.  0.003  YEAR  = 1,789 to 1,818  1,805  SALT MEAT  = 1 if contract was for salted beef, pork, or suet.  0.061  NOT FOOD  = 1 if contract was for non-food items, e.g., casks, hoops, etc.  0.022  FRESH  = 1 if contract was for fresh meat, bread, or vegetables.  0.127  INGREDIENTS  = 1 if contract was for flour, pease, oatmeal, vinegar, wheat.  0.282  Variable name  Definition  Mean  CONTRACT  = 0 if produced in-house, 1 if supply contract  0.782    = 2 if commission agent.    BISCUIT  = 1 if contract was for biscuit.  0.307  OXEN  = 1 if contract was for oxen.  0.009  BEER  = 1 if contract was for beer.  0.017  WATER  = 1 if contract was for water.  0.003  YEAR  = 1,789 to 1,818  1,805  SALT MEAT  = 1 if contract was for salted beef, pork, or suet.  0.061  NOT FOOD  = 1 if contract was for non-food items, e.g., casks, hoops, etc.  0.022  FRESH  = 1 if contract was for fresh meat, bread, or vegetables.  0.127  INGREDIENTS  = 1 if contract was for flour, pease, oatmeal, vinegar, wheat.  0.282  Table 2. Multinomial regressions for contract choice Variable  In-house production  Commission agent  (1)  (2)  (3)  (4)  BEER  5.34*  5.77*  −17.88  −18.04    (1.01)  (1.01)  (29,467)  (72,704)  OXEN  −35.03  −36.11  2.74*  4.38*    (1.6e+7)  (3.5e+7)  (0.28)  (0.40)  BISCUIT  −1.79*  −1.36*  −16.85  −16.22    (0.14)  (0.16)  (463)  (782)  WATER  −17.28  −17.75  −17.05  −16.39    (4,588)  (7,232)  (8,124)  (12,935)  SALT MEAT    2.90*    3.80*      (0.19)    (0.39)  NOT FOOD    24.58    1.47      (12,685)    (56,486)  FRESH    0.19    −0.58*      (0.14)    (0.30)  INGREDIENTS    0.80    2.02*      (0.12)    (0.32)  YEAR  −0.01*  −0.01  −0.11*  −0.09*    (0.006)  (0.007)  (0.01)  (0.01)  N  4,904  4,904  4,904  4,904  Log Likelihood  −2,210  −1,954  −2,210  −1,954  Pseudo R2  0.15  0.25  0.15  0.25  Variable  In-house production  Commission agent  (1)  (2)  (3)  (4)  BEER  5.34*  5.77*  −17.88  −18.04    (1.01)  (1.01)  (29,467)  (72,704)  OXEN  −35.03  −36.11  2.74*  4.38*    (1.6e+7)  (3.5e+7)  (0.28)  (0.40)  BISCUIT  −1.79*  −1.36*  −16.85  −16.22    (0.14)  (0.16)  (463)  (782)  WATER  −17.28  −17.75  −17.05  −16.39    (4,588)  (7,232)  (8,124)  (12,935)  SALT MEAT    2.90*    3.80*      (0.19)    (0.39)  NOT FOOD    24.58    1.47      (12,685)    (56,486)  FRESH    0.19    −0.58*      (0.14)    (0.30)  INGREDIENTS    0.80    2.02*      (0.12)    (0.32)  YEAR  −0.01*  −0.01  −0.11*  −0.09*    (0.006)  (0.007)  (0.01)  (0.01)  N  4,904  4,904  4,904  4,904  Log Likelihood  −2,210  −1,954  −2,210  −1,954  Pseudo R2  0.15  0.25  0.15  0.25  *Significant at the 1% level. Standard errors in parentheses. Reference Group: Supply Contracts. Table 2. Multinomial regressions for contract choice Variable  In-house production  Commission agent  (1)  (2)  (3)  (4)  BEER  5.34*  5.77*  −17.88  −18.04    (1.01)  (1.01)  (29,467)  (72,704)  OXEN  −35.03  −36.11  2.74*  4.38*    (1.6e+7)  (3.5e+7)  (0.28)  (0.40)  BISCUIT  −1.79*  −1.36*  −16.85  −16.22    (0.14)  (0.16)  (463)  (782)  WATER  −17.28  −17.75  −17.05  −16.39    (4,588)  (7,232)  (8,124)  (12,935)  SALT MEAT    2.90*    3.80*      (0.19)    (0.39)  NOT FOOD    24.58    1.47      (12,685)    (56,486)  FRESH    0.19    −0.58*      (0.14)    (0.30)  INGREDIENTS    0.80    2.02*      (0.12)    (0.32)  YEAR  −0.01*  −0.01  −0.11*  −0.09*    (0.006)  (0.007)  (0.01)  (0.01)  N  4,904  4,904  4,904  4,904  Log Likelihood  −2,210  −1,954  −2,210  −1,954  Pseudo R2  0.15  0.25  0.15  0.25  Variable  In-house production  Commission agent  (1)  (2)  (3)  (4)  BEER  5.34*  5.77*  −17.88  −18.04    (1.01)  (1.01)  (29,467)  (72,704)  OXEN  −35.03  −36.11  2.74*  4.38*    (1.6e+7)  (3.5e+7)  (0.28)  (0.40)  BISCUIT  −1.79*  −1.36*  −16.85  −16.22    (0.14)  (0.16)  (463)  (782)  WATER  −17.28  −17.75  −17.05  −16.39    (4,588)  (7,232)  (8,124)  (12,935)  SALT MEAT    2.90*    3.80*      (0.19)    (0.39)  NOT FOOD    24.58    1.47      (12,685)    (56,486)  FRESH    0.19    −0.58*      (0.14)    (0.30)  INGREDIENTS    0.80    2.02*      (0.12)    (0.32)  YEAR  −0.01*  −0.01  −0.11*  −0.09*    (0.006)  (0.007)  (0.01)  (0.01)  N  4,904  4,904  4,904  4,904  Log Likelihood  −2,210  −1,954  −2,210  −1,954  Pseudo R2  0.15  0.25  0.15  0.25  *Significant at the 1% level. Standard errors in parentheses. Reference Group: Supply Contracts. The results from table 2 strongly support the claim that the Victualling Board designed contracts in light of mitigating measurement costs. The variable BEER has a positive coefficient in regressions (1) and (2), but negative in regressions (3) and (4), reflecting that beer was almost exclusively produced in-house. The signs for the OXEN coefficients are the opposite: positive in regressions (3) and (4). Live beef was almost exclusively contracted for by commissioned agents. Finally, biscuit and water were contracted for by simple supply contracts.100 Table 2 also shows that salted meat was more likely produced in-house or by commissioned agent, and over time supply contracts became more common. 6.5 The End of Venal Clerks Just prior to the wars against France—and arguably during the Victualling Board’s finest hour—fifteen parliamentary commissions were established to examine “anachronisms” within most state offices, and notable for the Victualling Board was the Commission Appointed to Inquire into Fees, Gratuities, Perquisites and Emoluments, 1785–8.101 The ultimate report was the scathing Tenth Report of the Commissioners (1809) which focused specifically on the Victualling Board. One quote suffices to infer its tone: … it was not until we entered minutely into the investigation of the system of management in the Victualling Board, and the several Victualling Departments,… that we were convinced of the insufficiency of any partial alterations or amendments, and that nothing less than an entire new system would be likely to produce any effectual and permanent good…. [Tenth Report 1809, pp. 3–4] Clearly, at the time, the Victualling Board was viewed as an exemplar of managerial incompetence.102 One major concern of the Report was that bookkeeping accounts had been allowed to stockpile without being cleared, some for decades after “… the party had been dead many years prior to their being passed.”103 The failure to clear the accounts meant that no accounting verification had been done to ensure the fulfillment of the original contract, and the Commission worried that long unclosed accounts might open the door for fraud.104 The Tenth Report seems to refute the claim that the Victualling Board had been operating in a constrained efficient fashion.105 On the contrary, the Report reflects the improvements made in the ability to measure fundamentals over the eighteenth century. With this, the modern understanding of employment became more common, and the venal system came under attack, at the Victualling Board and other areas of the civil service. The reforms that took place during the Napoleonic wars created a modern bureaucracy among the various clerks of the Victualling Board by making them full time, fully salaried, with lines of authority, and measured qualifications.106 All of the language used in the Tenth Report speaks to the changed environment of modern measurement. The accounts should be “investigated” and “verified.” Agents should “view the operation,” have “regular and daily attendance” at manufacture sites, and have “practical skill and knowledge with respect to the particular species of manufacture.”107 The general language of the report for the various clerks and offices was to “adopt measures”, “examine,” “make inspection,” “accurately investigate,” have “cognizance” of the general operation, and “sign his initials… in proof of… investigation.”108 In addition, the Victualling Board was to introduce what today we would consider basic accounting and auditing functions, with standard accounting checks.109 All of this was to be monitored by committee of three who were to “minutely examine” all levels of operation, to consider the actual abilities of “officers, clerks, foremen, etc.” in order to remove those who were “old, infirm, and incapacitated,” and to conduct inspections “unexpectedly.”110 Furthermore, clerks and other officers were now expected to report to work at a specific time, and remain until four o’clock. Every single one of these activities, mundane in the modern world, rests on the ability to actually measure basic fundamental quantities. Several historians back up this measurement claim. Baker (1973, pp. 216–217) notes that from 1780 on the Victualling Board had a greater emphasis on “sounder information” and statistics, and that these internal changes made “… officials more accountable, more responsible, honest, and accessible…”. Macdonald (2010, p. 2) states that “… the accounting system for pursers and agents victualler changed from one of non-standardised paperwork to one of standard preprinted forms which had to be completed in a specific way.” Davey (2012, p. 159) claims that “Performance measurement was at the heart of the administrative changes.” And Wilkinson states: A detailed study of the Navy’s business after 1763 demonstrates problems and difficulties as one might expect but these were overcome. What emerges is an astonishing period of dynamic innovation, reform and change; a time of improvement and increased efficiency. In particular these years saw significant change in the civil and financial administration of the Navy… [2011, p. 3, (emphasis added)] These organizational changes created an early modern professional bureaucracy beneath the commissioners.111 As such, the new clerks and officers were to no longer receive any fees, gratuities, perquisites, or emoluments for services rendered. They were to be fully wage employees with superannuation, and no longer the owners of their office. These recommendations began to be implemented at the turn of the nineteenth century, with many in place by 1809, and they were completed in 1832 with the creation of the modern naval administration. The changing ability to monitor office functions lead to a movement from venal to professional office workers.112 6.6 Changes in the Variance of Average Victualling Costs Although the parliamentary reports culminated in the Tenth Report (1809), reforms began in ernest around 1801 with the arrival of the Earl of St. Vincent as the First Lord.113 In addition to the elimination of fees and perquisites, nominal salaries were increased and pensions were introduced.114 There were reforms in accounting, reporting, checking, monitoring of goods coming and going,115 and members of Parliament were no longer allowed an interest in government contracts.116 In short, between 1801 and 1809, after the measurement innovations, the office of the Victualling Board was converted into a modern bureaucratic organization based on measured inputs.117 In light of this, it is interesting to note what happened to the variance in average costs of operation during this time.118 For reference, figure 2 shows the navy’s real total costs over the 1793–1815 years with France. Aside from the low costs from the Treaty of Amiens (March 1802–May 1803), there is no systematic change in the real total costs over time.119 Figure 3, on the other hand, shows the real average victualling costs (total victualling costs per seaman) over the same period 17,931,815. Over the entire course of the war, the average cost of feeding a seaman increases—perhaps a reflection of changing operations over the war (both in the navy and army) and the rising real price of victuals. However, for matters considered here, the most significant feature is the change in the variance of average costs. Prior to the 1805 (the midpoint of reforms), the variance in the average cost of victualling is large relative to the post reform period. An F-test rejects the a null of equal variance (F = 12.36).120 Such a change is consistent with reforms built on a platform of measurement. Measurement allows predictability and stability in the delivery of service, which lowers the variance in the cost per seaman. Figure 2. View largeDownload slide Naval real total costs: 1793–1815 Figure 2. View largeDownload slide Naval real total costs: 1793–1815 Figure 3. View largeDownload slide Naval average total costs: 1793–1815 Figure 3. View largeDownload slide Naval average total costs: 1793–1815 7. Conclusion The British Victualling Board of the long eighteenth century may have a pop-culture reputation for incompetence and corruption, and the institutional devices they used can look “inefficient” when filtered through a modern lens. However, many modern historians have concluded that the Board was effective in accomplishing its task.121 Some conclude that it was also effective relative to other navies. Since all navies had access to the same technical elements of victual production, the source relative success was likely the organization elements of British victualling. Although little is systematically known about the victualling of other navies in Europe due to the destruction of many records, the evolution of victualling organization in France and Spain was often the opposite of that in England. For example, in the Spanish navies captains purchased their commissions and managed the victualling of the ship as a residual claimant—creating an incentive to minimize the cost of victuals on board.122 In France the captains also directly managed victuals. Rodger notes that “British naval victualling is a remarkable story of rising standards…. In France, by contrast, the story is of an inefficient, corrupt and costly system progressively disintegrating…”.123 An economic examination of the Victualling Board’s organization explains why it was an improvement over what preceded it, why it was so successful over the eighteenth century, and why it eventually changed. For the Victualling Board its organizational history was always a matter of choosing the “lesser of two weevils”; that is, choosing the best option in light of the pre-modern environment of poor measurement it operated in. Hence, the criticisms of the parliamentary reports resulted from examining the pre-modern institutional structure of the Victualling Board in the light of new modern methods of management, and there is no doubt that they were partly responsible for the changes that soon came. But the reports were a product of their time, and reflected the new modern world: when the measurement constraints changed, the Victualling Board organization changed with them, and this would have happened, reports or not. During the pre-modern period, when the measurement of fundamental quantities was too costly, the Board measured what it could (arrival times, gross quantities, and edible stores); it held inventories to ensure demands could always be met; it used the rule of three, mutual surveillance, in-house production and contract choice to mitigate transaction costs and police contractors; it used venal offices for its clerks to provide a private incentive to make sure accounts were eventually closed; and it altered the organization when office inputs became measurable at the end of the period. As such, the organizational history of the Victual Board provides another example of how the structure of property rights evolve based on changes in transaction costs.124 Acknowledgements Thanks to Yoram Barzel, Pete Leeson, Christoph Luelfesmann, Wing Suen, and two anonymous referees for comments; to naval historians Daniel Baugh, Aaron Graham, and especially Martin Wilcox for sharing comments and detailed knowledge of eighteenth century victualling; and to John Knowles for his comments and remembrance of O’Brien’s weevil pun. Conflict of interest statement. None declared. Footnotes 1 Masefield 1905, pp. 142–144. 2 See Knight and Wilcox (2010), or Macdonald (2010). 3 Pool, speaking of victualling during the Napoleonic Wars agrees: “The fact is that the country could hardly have survived the war if the arrangements for providing the ships and supplying them with stores had not been basically sound and well-administered.” (1966, p. 130). 4 The Western Squadron continuously sat off the western French shores attempting to blockade enemy harbors, intercept outbound convoys, reduce the need for strong fleets in the Mediterranean, Caribbean, and East Indies, eliminate the need for a channel fleet because the French lacked a safe harbor along the channel, and shield British merchant ships (Baugh 2004, pp. 250–251). North et al. (2009, pp. 181–182) credit the Western Squadron as a major factor in eighteenth century British military success. Having a ship at sea for months on end was cheaper than alternating vessels, and ships tended to become healthier and more effective the longer they remained at sea. Crews benefited from the constant sail-handling and drill, were not exposed to the diseases and dubious pleasures of the port towns, and had fewer opportunities to desert. 5 Knight and Wilcox (2010, p. 52) claim that 400,000 is a “conservative estimate” of the total number of men fed by the victualling board in 1801. Most of these men would have been in the army, and the number also includes some non-British troops. About 140,000 were actual seamen. Macdonald (2010, p. 2) claims that at its peak the victualling board fed up to 250,000 men. Obviously earlier counts would have been smaller. No matter the exact number, throughout the era it was a monumental task of organization to manage. 6 Davey 2012, p. 164; Baugh 2004, p. 240. This held even for the remote East Indies station (Knight and Wilcox 2010, p. 171). This is not to say the service was perfect. An ironic example arises from the battle of Trafalgar. The combined French and Spanish fleet had been eluding Nelson for months, and were partly drawn out of Cadiz that fateful October day in 1805 because Nelson had to send six ships off for water and provisions, leaving his squadron with an inferior number (Davey 2012, p. 51; Rodger 2004, p. 537). 7 Rodger 2004, p. 486. There is a general consensus among historians that there was no discrete change, but rather a continued improvement over time (see e.g., Baugh 2004, p. 256; Baugh 1965, pp. 374–375; Knight 2008, p. 143; Rodger 1986, pp. 485–487; or Bowen 2013, p. 242). 8 Naval historian Daniel Baugh finds it difficult to underestimate the success of the Victualling Board: “… ignorantly criticised in the popular historical literature yet probably the most important triumph of the eighteenth century British naval administration: the victualling service” (2004, p. 256). Knight and Wilcox (2010) or Buchet (2013) also look favorably on the Victualling Board. For most of the historical literature, however, there’s no explanation of why the organizational structure actually worked. 9 Graham (2013, p. 807), regarding pre-modern supervisory abilibites, notes that “the various naval boards were incapable of [directly] detecting corruption and misappropriation among their own officials, let alone their contractors.” 10 Davey 2012, p. 38. 11 To my knowledge, the only economists to have considered British victualling are North, Wallis, and Weingast (2009, pp. 181–187). Their analysis provides an example of their general argument on social orders, and they see the success of the Victualling Board as arising from (1) the Navy’s ability to credibly issue bills to finance operations, (2) the development of elite impersonal suppliers, and (3) the Victualling Board’s status as an ongoing impersonal institution. My analysis is consistent with the North, Wallis, Weingast broad claim; however, I explain the Board’s solutions in more detail. For example, here the key institutional changes that allowed for trustworthy debt instruments is identified, I explain why debt issued by the Victualling Board in the form of bills was instrumental in ensuring that victuals arrived in time, and I explain how the old world institutions of patronage and venality worked during the ultimate transition to impersonal market institutions. 12 See for example, Geddes and Lueck (2002) on female enfranchisement, Reed and Bekar (2003) on usury, Barzel (1977) on slavery, Greif (2006) on medieval trade, or Allen (2002) on the British Navy. 13 See for example North and Thomas (1973) or McCloskey (1976). The claim of first-best efficiency led to various debates. For example, Fenoaltea (1975) challenged the North and Thomas claim in the context of the Manorial system. 14 See for example, Klein and Ogilvie (2016) who see rent extraction as the best interpretation of serfdom. 15 For a book length treatment, see Greif (2006), for an application to dueling, see Allen and Reed (2006). 16 See Allen (2002), Benjamin and Tifrea (2007), and Benjamin and Thornberg (2007) for evidence of this hypothesis among the officers and crew. 17 This literature is vast. See Ogilvie (2007) for a terse survey. 18 The term “victuals” refers to food, but the Victualling Board also handled other provisions such as kindling, casks, candles, flags, staves, and hoops. 19 Pool 1966, p. ; Oppenheim 1961, p. 103. 20 Conway et al. (2013, p. 253). 21 See Rodger 1997, pp. 364–378 for an extensive set of examples. 22 Rodger 2004, p. 105–106, Oppenheim 1961, p. 140. 23 Rodger 2004, p. 42. 24 Rodger 1997, p. 367. See also North et al. (2009, pp. 182–183), or Conway et al. (2013, pp. 252–253). 25 Rodger 2004, p. 106. 26 Baugh 1965, pp. 373–374; Rodger 2004, pp. 26–27, p. 43, p. 69; Buchet 2013, p. 16. See also Buchet (2013, pp. 16–24) for examples of poor sixteenth century incentive structures that lead to poor provisions and high rates of ship mortality. 27 Rodger 2004, p. 143. 28 Quoted in Rodger 2004, p. 193. 29 Rodger 2004, p. 144. 30 One might think that such a watershed institutional moment was the work of Samuel Pepys. Ironically, Pepys had been ousted from naval administration in 1679 and did not take part in the Admiralty Commission that established the new structure in 1683. Once back in power in 1684 under the new, and short-lived, title of “secretary for the affairs of the admiralty of England,” he retained the new system—although he forever destroyed the reputations of the commissioners, and no doubt influenced future popular opinion of them. See Davies (1989) for a detailed account. 31 Macdonald (2010) and Knight and Wilcox (2010) provide book length treatments of the day-to-day operations of the Victualling Board. Syrett (1996), in describing an infamous victualling scandal after the American war of independence, also provides a detailed account of daily activity. 32 Baugh 2015, p. 24. 33 The Crown had always used debt (bills) to finance naval operations. However, as North et al. (2009, pp. 183–186) point out, these bills always had a “personal” element, and therefore, were not always paid back in time or in order. 34 Davey 2012, p. 38. 35 Pool 1966, p. 139; Pool 1964, p. 163. 36 Morriss 2011, p. 280; Davey 2014, p. 30. 37 Macdonald 2010, p. 113. 38 Davey 2014, pp. 32–35. 39 “Venality” in the pre-modern era was a matter of degree, and at the Victualling Board, it may be better to say that many offices had a “venal character.” “Customary transfer fees” were made to acquire an office, “poundage” was paid on contract fulfillment, and merit played some role. Here, for simplicity, I will refer to such offices as “venal.” 40 A charge and discharge system reconciles all distributions of money going in and out of an account. It is simpler than double entry bookkeeping, but is hard to use for monitoring because a reckoning only happens once the account is closed. Double entry bookkeeping was not introduced until 1832 (Mann 2014). 41 See Mann, Chapter 2, and the Fifth Commission of Public Accounts, 1781, for a detailed description of this practice. The ability to retain funds could be significant. To take a naval example outside the Victualling Board, over the period 1757–1777 the paymaster of marines, John Tucker, had over £2.25 million pass through his accounts (Morriss 2011, p. 123). 42 Knight and Wilcox 2010, p. 29; Davey 2012, p. 149. The entire British state had 24,598 employees in 1815 (Morriss 2011, p. 9). 43 Knight and Wilcox 2010, p. 4. 44 Disease, especially typhus and scurvy, were perpetual threats in the age of sail that depended on the quality and timing of victuals. The Victualling Board managed scurvy with the innovation of transferring live animals and fresh produce on board reduced the problems. By the end of the eighteenth century scurvy was effectively eliminated (Rodger 2004, pp. 485–487; Knight 2008, p. 143; Baugh 2004, pp. 250–251). 45 This short section is based on Chapter 2 of Allen (2012), and Glennie and Thrift (2009). 46 “People could take precision or leave it, and often they left it.” Glennie and Thrift 2009, p. 276. 47 Glennie and Thrift 2009, p. 293. 48 See Landes (1983) for another history of time keeping, and the generations it took to adopt “clock-time.” 49 Baker 1973, p. 217. Indeed, Allen (2012) argues these measurement changes led to a general “institutional revolution” and the emergence of the modern world. 50 Rodger 2004, p. 43. 51 Baugh (2011, pp. 329–330) provides a detailed example of typical hazards in shipping. Reporting French efforts to supply Quebec during the Seven Years War, he notes that for one convoy eight ships left Bordeaux, two were captured and six reached Quebec. Of the six that returned, two were captured, one sank, one returned extremely damaged by battle, and two returned unscathed. 52 Formally, let this cumulative density be given by F(a|e), where a is the known arrival date, e is the effort by the Victualling Board, and this density has first-order stochastic dominance in e. For the Victualling Board the payoff (π) objective would be:   π=∫0a⁎π(a)f(a|e)da−P(1−F(a⁎|e))−c(e),where P is the penalty when supplies arrive too late, and c the cost of effort. Given the nature of the penalty, the optimal solution is to choose a very high value of e in order to minimize the probability of arriving late, 1−F(a|e). 53 Such malfeasance are examples of transaction costs, which depend, in part, on the variability of Nature. Hence, the greater the role of variability, ceteris paribus, the greater are the potential transaction cost problems. See Allen (1991, 2000, 2015) for the definition and conditions for transaction costs. 54 The reciprocal case would be a ship arriving on a specific date where the stores are located. 55 This is not to say that the seamen were under fed. Compared to their land side contemporaries, seaman caloric intake was higher and involved larger portions of proteins (Rodger 1986, p. 87). 56 It is beyond the scope of this paper to address the question of how the Admiralty would ever know that yards were stocked, a contracted number of victuals delivered, or if sailors could not stomach food. As explained in Allen (2005, 2012), the solution was a system that relied on “watchdogs” and “mutual surveillance,” combined with enormous penalties if caught for misconduct. 57 See Baugh 1965, p. 373–390, for a general discussion of the Declaration. 58 Baugh 1965, p. 373–374. 59 Mann 2014, pp. 241–242. 60 Baugh 1965, p. 389. 61 Baugh 1965, p. 390 and p. 446. The actual level of stores is unknown, but the reputation of the Victualling Board was that their inventories were “excessive.” 62 Baugh 1965, p. 432. 63 Buchet 2013, p. 57. Earlier, during the sixteenth century, victualling failures prevented the fleet from sustaining a blockade of the smaller Dutch coast (Rodger 2004, p. 43). 64 Buchet 2013, pp. 34–43. Baugh (2015, p. 39) notes that during the eighteenth century “Parliament was mainly interested in the navy’s readiness and performance.” 65 Buchet 2013, pp. 255–256. They were also uncommon during the Napoleonic Wars. Writing regarding that time, Davey (2009, p. 256) notes “The system responded well to changing demands and circumstances…. The system was flexible enough to account for changes in the quantity of provisions needed.” 66 Roseveare (1969, p. 93), in this context notes “… there remains an aspect of military expenditure which completely defied the principle of parliamentary sanction and which was barely checked by Treasure restraint. It was the freedom of the Navy… to incur debts far beyond the Estimates…”. 67 Baugh 2015, p. 39; see also Mann 2014, p. 223. 68 One might think that a moral hazard problem might arise in the Victual Board’s issuance of too many bills. Here is where the patronage system was binding. There was a day of reckoning when the accounts were closed, and at some point such behavior would be discovered and a commissioner(s) could have their office taken away. 69 For example, Mann (2014, p. 245) mistakenly claims that the issuing of debt in this manner “is a startling example of corruption at the highest levels.” 70 In a patronage system the reputation (sunk investment) of the individual trumped other considerations. This was certainly the case with the Victualling Board. Baugh (1965, p. 60), notes “Although many Commissioners were competent men of business,… their previous experience in victualling matters was nil…”. At the same time, some attention was paid towards merit, experience, knowledge, and talent. The concept of an “expert,” however, was not used as a noun in the eighteenth century (Davey 2014, p. 31). 71 This was a complicated social arrangement, the details of which, are beyond the scope of this paper. See Allen (2012) for elaboration. 72 Macdonald 2010, p. 163. 73 Pool 1966, p. 113. 74 See Davey (2014, pp. 30–31) for a discussion of victual patronage based on the correspondence of First Lord Mulgrave. 75 Buchet 2013, p. 18. 76 The closing of accounts could take some time as knowledge over the performance markers was acquired and verified. Still most of the accounts over the life of the Board were passed and liquidated within a reasonable period of time that allowed for a verification of the arrival date, quantity, and minimal quality required. 77 Which did happen. See Syrett (1996) for an infamous case. 78 The lack of auditing extended beyond the books. It was not until 1749, sixty-six years after the creation of the naval boards, that the admiralty conducted its first inspection of the dockyards (Middleton 1991, p. 21). It would be 1764 before the second visit took place (Wilkinson 1998, p. 418). 79 Graham 2013, p. 810. Graham (2013) provides a fascinating account of the exploits of James Brydges, paymaster of the Forces Abroad, 1705–1713. Though he was able to direct large sums to himself, he was well aware that until his accounts were cleared, the “axe might fall” (p. 837). 80 In discussing public “auditing” during the long eighteenth century, Graham (2013, p. 810) notes that this form of mutual surveillance was one of two cornerstones. The other was personal liability. 81 See Allen 2012, Chapter 5, for detailed examples of this in the Navy. Brewer (1988, pp. 102–09) provides another example of a rule of three from eighteenth century tax collection. 82 Syrett 1996, p. 129; Macdonald 2010, pp. 115–120; Morriss 2011, p. 118. 83 Davey 2012, p. 157. 84 Macdonald 2010, p. 201. 85 Macdonald 2010, pp. 76–77. 86 Macdonald 2010, p.164. 87 Morriss 2011, pp. 303–307. 88 Macdonald 2010, p. 61. 89 Syrett 1996, p. 131. 90 Knight and Wilcox 2010, p. 120. 91 Knight and Wilcox 2010, p. 94, state “contracts became increasingly standard and formalised as time went on.” 92 See Wilcox (2010) for details about the data. 93 Sumner 2013, p. 43. 94 Knight and Wilcox 2010, p. 125. 95 Knight and Wilcox (2010, p. 99) discuss an interesting case in which a captain returned some bad beef, which then triggered an inquiry with the contractor. In a letter to the board the contractor stated “… I would not suffer Cow or Bull beef to be sent on board one of His Majesty’s Ships,…”. Clearly, the contractor knew the importance of oxen. 96 Barzel (1982) provides the seminal argument for vertical integration based on measurement costs. 97 For example, Knight and Wilcox 2010, p. 85 and pp. 97–101. 98 John Stuart Mill, quoting Charles Babbage, understood this: The difficulty and expense of verification are in some instances so great, as to justify the deviation from well-established principles. Thus, it is a general maxim that Government can purchase any article at a cheaper rate than that at which they can manufacture it themselves. But it has, nevertheless, been considered more economical to build extensive flour-mills (such as those at Deptford), and to grind their own corn, than to verify each sack of purchased flour, and to employ persons in devising methods of detecting the new modes of adulteration which might be continually resorted to. [Mill 1899, vol. i p. 137] 99 Hence a positive coefficient implies the variable is positively correlated with supply contracts. 100 This confirms the claim by Buchet 2013, p. 226. 101 Morriss 2011, p. 125. 102 One modern historian, judging the victualling commissioners with a modern lens, agrees with the Parliamentary reports. Macdonald (2010) noted that victualling commissioner attempts at monitoring were “notable for their naivety” (p. 55), that “they might as well not have bothered” to obtain “accurate figures” (p. 56), and that they “failed” to maintain any proper observation of their employees (pp. 185–186). 103 Tenth Report, 1809, pp. 5–7. With hindsight, the Commissioner’s findings were not representative. Most of the uncleared accounts had arisen over the massive growth of naval activity that took place during the wars with France—the very moment of the inquiry. 104 In the 1781 Third Report, it was also found that a 1689 account still had a balance of £27,611 outstanding (Reports 1781, p. 27). 105 Although the Tenth Report was apoplectic at the accounting inattention, the Report actually found no cases of verified malfeasance. After noting the possibility of all sorts of wrongdoing in the victual operations, the report states: We do not mean, however, to throw out suspicions of such having been the case; on the contrary, we will state distinctly that we should not be authorized, by any thing which has come before us, to convey such an insinuation; [1809, p. 15] 106 Parris 1969, pp. 22–24; Pool 1966, p. 114. 107 Report (1809, p. 14). 108 Report (1809, p. 20). 109 Report (1809, p. 22). 110 Report (1809, p. 23). 111 Further evidence of the world of measurement comes from a 1796 Order in Council that brought Samuel Bentham into the naval administration to reform it along modern lines. He had a focus on “individual responsibility”, and introduced new management positions that inspected supplies to actual specifications (Pool 1966, pp. 116–119). With respect to clerks he “… established a system of measurement, quantifying and paying for piecework on the spot within the week.” (Morriss 2011, p. 19). 112 The patroned commissioners would hang on until the Victualling Board was dissolved in 1832 along with other similar offices in the navy. Allen (2002) argues that the advent of steam propulsion, the common use of the chronometer, and the ability to directly monitor the day-to-day victual operations, allowed for direct monitoring of the victual administrators. Hence the commissioners became fully salaried professional bureaucrats in 1832. 113 “St. Vincent was prone to mistake slackness and incompetence for corruption and crime,” (Pool 1966, p. 118), but the result was a zealot for reform. 114 Macdonald 2010, pp. 39, 108–126; Baker 1973, p.217. Morriss (2011, p. 17) refers to this as a bureaucratic “silent revolution.” 115 Davey 2012, p. 164. 116 Pool 1964, p. 162. 117 Knight 2008, p. 171. 118 Data on costs come from the appendices of Knight and Wilcox (2010), and supplemented by appendix vi in Rodger (2004). 119 They are slightly rising, but controlling for the treaty, the time trend is statistically insignificant. 120 Dropping the years 1802 and 1803 yields the same result, with F = 8.96. 121 This is the basic conclusion of Knight (2008, p. 143), Knight and Wilcox (2010), Pool (1966, p. 130), Wilkinson (2004, p. 7), and Davey (2012, p. 164). 122 Rodger 1986, p. 107. 123 Rodger 2004, pp. 306–07. In perhaps the most comprehensive study of the Victualling Board, Knight and Wilcox conclude: Their administrative acumen and their effective engagement and management of countless contractors gave Great Britain the decisive edge in the Great Wars with France. [p. 214, 2010] 124 Another confirmation of Coase (1960), and in this specific case, Allen (2012). References Allen, D.W. ( 1991). What are transaction costs? Research in Law and Economics  14, pp. 1– 18. Google Scholar CrossRef Search ADS   Allen, D.W. ( 2000). Transaction costs. In Bouckaert Boudewijn and De Geest Gerrit (eds), Encyclopedia of Law and Economics (Volume One: The History and Methodology of Law and Economics) . Chelthenham: Edward Elgar Press, pp. 893– 926. Allen, D.W. ( 2002). The British navy rules: monitoring and incompatible incentives in the age of fighting sail. Explorations in Economic History  39, pp. 204– 31. Google Scholar CrossRef Search ADS   Allen, D.W. ( 2005). Purchase, patronage, and professions: incentives and the evolution of public offiffice in pre-modern Britain. Journal of Institutional and Theoretical Economics  161, pp. 57– 79. Google Scholar CrossRef Search ADS   Allen, D.W. ( 2012). The Institutional Revolution: Measurement and the Economic Emergence of the Modern World . Chicago: University of Chicago Press. Allen, D. and Reed, C. ( 2006). The Duel of Honor: Screening for Unobservable Social Capital. American Law and Economics Review  8( 1), pp. 81– 115. Google Scholar CrossRef Search ADS   Baker, N. ( 1973). Changing attitudes towards government in eighteenth century Britain. In Whiteman A., Bromley J.S. and Dickson P.G.M. (eds), Statesmen, Scholars, and Merchants . Oxford: Clarendon Press, pp. 203– 219. Barzel, Y. ( 1977). An Economic Analysis of Slavery. Journal of Law and Economics  20( 1), pp. 87– 110. Google Scholar CrossRef Search ADS   Barzel, Y. ( 1982). Measurement cost and the organization of markets. The Journal of Law & Economics  25, pp. 27– 48. Google Scholar CrossRef Search ADS   Baugh, D. ( 1965). British Naval Administration in the Age of Walpole . Princeton: Princeton University Press. Baugh, D. ( 2004). Naval power: what gave the British navy superiority? In Escosura L. (ed), Exceptionalism and Industrialisation: Britain and its European Rivals, 1688–1815 . Cambridge: Cambridge University Press, pp. 235– 258. Google Scholar CrossRef Search ADS   Baugh, D. ( 2011). Global Seven Years War: 1754–1763 . Harlow: Pearson Education. Baugh, D. ( 2015). Parliament, naval spending and the public: contrasting financial legacies of two exhausting wars, 1689–1713. Histoire & Mesure  XXX-2, pp. 23– 50. Google Scholar CrossRef Search ADS   Bowen, H.V. ( 2013). The contractor state, c. 1650–1815. International Journal of Maritime History  25, pp. 239– 74. Google Scholar CrossRef Search ADS   Benjamin, D.K. and Thornberg, C. ( 2007). Organization and incentives in the age of sail. Explorations in Economic History  44, pp. 317– 41. Google Scholar CrossRef Search ADS   Benjamin, D.K. and Tifrea, A. ( 2007). Learning by dying: combat performance in the age of sail. The Journal of Economic History  67, pp. 968– 1000. Google Scholar CrossRef Search ADS   Brewer, J. ( 1988). The Sinews of Power: War, Money and the English State, 1688–1783 . Cambridge: Harvard University Press. Buchet, C. ( 2013). The British Navy, Economy and Society in the Seven Years War . Woodbridge: The Boydell Press. Coase, R. ( 1960). The problem of social cost. Journal of Law and Economics  3, pp. 1– 44. Google Scholar CrossRef Search ADS   Conway, S., Harding, R., Harding, P. and Helen, J. ( 2013). Eighteenth-century Britain: the quintessential ‘contractor state’. International Journal of Maritime History  25, pp. 248– 53. Davies, J.D. ( 1989). Pepys and the admiralty commission of 1679–84. Historical Research  62, pp. 34– 53. Google Scholar CrossRef Search ADS   Davey, J. ( 2009). Within hostile shores: victualling the royal navy in European waters during the French revolutionary and Napoleonic Wars. International Journal of Maritime History  21, pp. 241– 60. Google Scholar CrossRef Search ADS   Davey, J. ( 2012). The Transformation of British Naval Strategy: Seapower and Supply in Northern Europe, 1808–1812 . Woodbridge: Boydell Press. Davey, J. ( 2014). Expertise and naval administration: Lord Mulgrave and the victualling board, 180710. Journal of Maritime Research  16, pp. 29– 41. Google Scholar CrossRef Search ADS   Fenoaltea, S. ( 1975). The rise and fall of a theoretical model: the manorial system. Journal of Economic History  35, pp. 386– 409. Google Scholar CrossRef Search ADS   Geddes, R. and Lueck, D. ( 2002). The gains from self-ownership and the expansion of women’s rights. The American Economic Review  92, pp. 1079– 92. Google Scholar CrossRef Search ADS   Glennie, P and Thrift, N. ( 2009). Shaping the Day: A History of timekeeping in England and Wales 13001800 . Oxford: Oxford University Press. Google Scholar CrossRef Search ADS   Graham, A. ( 2013). Auditing Leviathan: corruption and state formation in early eighteenth-century Britain. English Historical Review  CXXVIII, pp. 806– 38. Google Scholar CrossRef Search ADS   Greif, A. ( 2006). Instituions and the Path to the Modern Economy: Lessons from Medieval Trade . Cambridge: Cambridge University Press. Google Scholar CrossRef Search ADS   Klein, A. and Ogilvie, S. ( 2016). Occupational structure in the Czech lands under the second serfdom. Economic History Review  69, pp. 493– 521. Google Scholar CrossRef Search ADS   Knight, R. ( 2008). Politics and trust in victualling the navy, 1793–1815. The Mariner’s Mirror  94, pp. 133– 49. Google Scholar CrossRef Search ADS   Knight, R. and Wilcox, M. ( 2010). Sustaining the Fleet, 1793–1815: War, The British Navy, and the Contractor State . Woodbridge: The Boydell Press. Landes, D. ( 1983). Revolution In Time: Clocks and the Making of the Modern World . Cambridge: Harvard University Press. Macdonald, J. ( 2010). British Navy’s Victualling Board, 1793–1815: Management Competence and Incompetence . Woodbridge: The Boydell Press. Mann, I. ( 2014). An Italian in the Royal Navy: The Recruitment of Double Entry Bookkeeping. Phd. thesis, University of Wollongong. Masefield, J. ( 1905). Sea life In Nelson’s Navy . London: Methuen & Co. McCloskey, D. ( 1976). English Open Fields as Behavior Towards Risk. Research in Economic History, Fall. Greenwich, CT: JAI Press. Middleton, R. ( 1991). The visitation of the royal dockyards, 1749. The Manner’s Mirror  77, pp. 21– 30. Google Scholar CrossRef Search ADS   Mill, J.S. ( 1899). Principles of Political Economy . New York: Colonial Press. Morriss, R. ( 2011). The Foundations of British Maritime Ascendancy: Resources, Logistics and the State, 1755–1815 . Cambridge: Cambridge University Press. North, D.C and Thomas, R. ( 1971). The rise and fall of the manorial system. Journal of Economic History  31, pp. 777– 803. Google Scholar CrossRef Search ADS   North, D. and Thomas, R. ( 1973). The Rise of the Western World: A New Economic History . New York: Cambridge University Press. Google Scholar CrossRef Search ADS   North, D.C, Wallis, J.J. and Weingast, B.R. ( 2009). Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History . Cambridge: Cambridge University Press. Google Scholar CrossRef Search ADS   Ogilvie, S. ( 2007). Whatever is, is right? Economic institutions in pre-industrial Europe. Economic History Review  60, pp. 649– 84. Google Scholar CrossRef Search ADS   Oppenheim, M. ( 1961). A History of the Administration of the Royal Navy and of Merchant Shipping in Relation to the Navy . London: Bodley Head. Parris, H. ( 1969). Constitutional Bureaucracy: The Development of British Central Administration Since the Eighteenth Century . New York: Kelley. Pool, B. ( 1964). Navy contracts in the last years of the navy board (1780-1832). Mariner’s Mirror  50( 3), pp. 161– 76. Google Scholar CrossRef Search ADS   Pool, B. ( 1966). Navy Board Contracts: 1660–1832 . Hamden: Archon Books. Reed, C. and Bekar, C. ( 2003). Religious prohibitions against usury. Explorations in Economic History  40, pp. 347– 68. Google Scholar CrossRef Search ADS   ( 1781). Reports of the Commissioners on Public Accounts Third Report. Rodger, N.A.M. ( 1986). The Wooden World: An Anatomy of the Georgian Navy . New York: Norton. Rodger, N.A.M. ( 1997). The Safeguard of the Sea: A Naval History of Britain . New York: Norton. Rodger, N.A.M. ( 2004). The Command of the Ocean: A Naval History of Britain, 1649–1815 . London: Norton. Roseveare, H. ( 1969). The Treasury: The Evolution of a British Institution . New York: Columbia U. Press. Syrett, D. ( 1996). Christopher Atkinson and the victualling board, 1775–82. Historical Research  69, pp. 129– 42. Google Scholar CrossRef Search ADS   Sumner, J. ( 2013). Brewing Science, Technology and Print, 1700–1889 . London: Pickering & Chatto. Tenth Report of the Commissioners for Revising and Digesting the Civil Affairs of His Majesty’s Navy, Commons reports 11th April 1809. Wilcox, M. ( 2010). Supporting Documentation for Sustaining the Fleet: Victualling Contracts and Contractors 1793–1815. Greenwich Maritime Institute. Wilkinson ( 1998). The earl of egmont and the navy, 1783–6. The Mariner’s Mirror  84, pp. 418– 33. CrossRef Search ADS   Wilkinson ( 2004). The British Navy and the State in the 18th Century . Woodbridge: Boydell Press. © The Author 2017. Published by Oxford University Press on behalf of the European Historical Economics Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
TI - “The Lesser of Two Weevils”: British victualling organization in the long eighteenth century
JF - The European Review of Economic History
DO - 10.1093/ereh/hex020
DA - 2018-05-01
UR - https://www.deepdyve.com/lp/oxford-university-press/the-lesser-of-two-weevils-british-victualling-organization-in-the-long-x5o26uhyIJ
SP - 233
EP - 259
VL - 22
IS - 2
DP - DeepDyve
ER -