THE SEVENTEENTH-CENTURY PUZZLE OF TAX FARM LEASE PRICES

France did not emerge whole from the Middle Ages, but grew gradually through a process of accretion. The early-seventeenth-century fiscal system reflected this fact. Most taxes began as necessary expedients to finance wars and then, over time, became permanent. Few taxes were untouched by the implicit bargains made between the crown and the various regions being brought into the fold. Sometimes exceptions were conceded to a privileged group, such as the nobility or the clergy, others were made to entire geographic regions.⁹

By the sixteenth century the organization of tax collection and the institutions in which these organizations were embedded reflected this piecemeal accumulation.

There were two primary types of taxes. The tailles were impôts de répartition. The crown decided on an amount to be collected each year and then divided this amount among the collectors, who were usually given a percentage of the take as compensation. The répartition was typically handled at the level of the généralité (roughly equivalent to states) and then local officials broke down the burden of the tax to parish-level collectors. The second method of collection, done through the tax farms, was an impôt de quantité. Tax farms had specific rates associated with transactions. In 1600 there were more than ten large tax farms in France, which were allocated using competitive English (ascending bid) auctions every four to six years.¹⁰

The main taxes collected through the farms were the gabelles (salt tax), the traites (internal and external customs duties), the aides (primarily wholesale and retail taxes on wine and spirits), and the domaines (traditional seigniorial dues).

There is a puzzle surrounding the evolution of the tax farm system during the seventeenth century. Figure 1 shows real revenues collected through the tax farms. After the civil disturbances known as the Fronde (1648–1653) there was a significant increase in revenues through the system. By the most conservative estimate, real receipts during the second half of the century were triple those of the first half.¹¹

Between 1662 and 1773 tax farm revenues accounted for an average of 45 percent of total revenues.¹² By itself, this increase is not necessarily surprising. Tax farm revenues grew at about the same rate between 1661 and 1700 as they had between 1600 and 1640. What makes these increases puzzling is that, at precisely the time the crown received more from the tax farm system, it also abandoned the practice of competitively auctioning tax farm contracts. Table 2 illustrates that the proportion of revenues collected through the single largest and two largest tax farms dramatically increased during the seventeenth century. By 1700 the tax farm system was virtually a monopsony. Should not the crown have received less revenues through the farms after abandoning competitive auctions? What could the state possibly gain by encouraging the creation of a powerful cartel of tax collectors?

One obvious explanation for this puzzle is that real output dramatically increased around 1661. According to most measures, however, real output either stagnated, or declined, during the second half of the seventeenth century. This is so well accepted that historians call this time the “longue période des difficultés.”¹³

There was a reduction in agricultural and industrial production, decreases in the productivity of agriculture, and a steady decline in the price level.¹⁴

It is more difficult to deal with the question of what was happening to tax rates at the time of the increase in farm revenues. On the one hand, Louis XIV's Finance Minister, Colbert, reduced the amount of the taille by half between 1656 and 1664. This certainly reduced the cost of collecting taxes through the farms. It is unlikely, however, that this accounted for a tripling of revenues, especially as Colbert increased the tailles to their previous level after 1672.¹⁵

Perhaps the tax rates associated with the farms increased during this period? There were literally hundreds of different tax rates in France during the seventeenth century and calculating something like an average tax rate is impossible. However, most tax farm receipts were accounted for by three main taxes, the gabelles, the traites, and the aides.¹⁶

Changes in the tax rate for the gabelles are relatively easy to identify because it was collected by selling salt at an above market price. Furthermore, Emmanuel le Roy Ladurie has compiled price and quantity data for the Gabelles de France during the seventeenth century, thus allowing us to deduce something about tax rates. Between 1600 and 1624 the average price for a quintal of salt was 150 livres tournois. Between 1640 and 1650, well after Cardinal Richelieu's tax rate increases (the tour de vis) the average price was about 400 livres tournois. Receipts from the Gabelles de France increased from 1,655,453 livres tournois to about 3,952,521 livres tournois during the same period. This implies that the elasticity of tax revenues with respect to the tax rate on salt was about 0.86.¹⁷

A 10 percent increase in the tax rate could be expected to yield an 8.6 percent increase in tax revenues. Between 1660 and 1680 the Gabelles de France brought in an average of 6,146,967 livres tournois a year. Assuming elasticity did not change after 1660, this implies a 42 percent increase in the tax rate in order to explain the increased revenues. However, the price of a quintal of salt actually declined a small amount between 1660 and 1680!¹⁸

The traites were composed of a mishmash of internal and external tariffs. The largest set of internal tariffs was known as the Cinq Grosses Fermes and constituted a huge tariff boundary surrounding the center of France. In an ordonnance of September 1664 Colbert simplified and lowered the tariffs associated with the Cinq Grosses Fermes.¹⁹

Nonetheless, between 1664 and 1682 the revenues from the Cinq Grosses Fermes increased from 9,000,000 to 11,830,000 livres. It is possible, that the tax farmers ignored the royal ordonnance and continued to tax at the same rates, or even increased them. However, an increase in the rates of internal tariffs should also have increased the price of traded goods. Among the most important of these goods was grain, however, its price fell approximately 41 percent between the periods 1646–1665 and 1666–1685.²⁰ As a mercantilist, Colbert did increase at least one component of the traites. In 1667 he doubled the tax rate on foreign trade.²¹

The aides were, by far, the most chaotic of the tax farms in 1661. They were composed of wholesale and retail taxes on virtually every good the crown could define. Between 1662 and 1683 the amount collected through the aides increased by about 170 percent. There is no ordonnance mentioning an increase in rates, however, until 1674, well after the lion's share of the increase had occurred.²²

Despite this fact, it is likely that tax rates associated with the aides were better enforced and defined after 1661. This would, indeed, correspond to an increase in tax rates, however, I will argue that it was also symptomatic of a larger trend in the tax farm system that accompanied unification. Namely, that the consolidation of the tax farms after 1661 resulted in efficiencies in tax collection, such as more clearly defined jurisdictions for the farmers and less replication of offices across farms.²³

Between 1600 and 1661 consecutive finance ministers attempted to take advantage of these economies of scale by consolidating leases. However, a consolidated lease required greater capital investment from the farmer who bid on it. Existing institutions that encouraged competition amongst the farmers also raised their costs of pooling capital to bid on leases. This, combined with the fact that much of the capital that was invested in a farm was paid up front to the crown as loans, resulted in powerful incentives for the king to violate the farmers' property rights by refusing to pay back loans or otherwise altering the terms of their contract. After 1661 institutions evolved such that it was possible for a powerful cartel of tax farmers emerge. This cartel, known as the Company of General Farms, was able to pool enough capital so that it could take advantage of the scale economies associated with tax collection. Furthermore, they used this capital to become a vital financial intermediary for the crown. As a result, the king was much less likely to violate their property rights by, for example, defaulting on a loan or reselling one of their taxes to another collector.

A TRANSACTION COST EXPLANATION OF TAX FARMING

Risk aversion cannot explain the evolution of the tax farm system during the seventeenth century. Namely, that it went from a decentralized to a centralized organization and that this transformation was accompanied by an increase in tax revenues from the farms that was not simply due to changing output, prices, or tax rates. An alternative explanation, which can successfully account for the increase in lease prices, is based on transaction cost economics. When tax farming was first introduced during the Hundred Years War, the farms were relatively small and required little capital on the part of the tax farmer. By the beginning of the seventeenth century, for reasons that will be discussed, the amount of investment required to effectively manage a farm increased significantly. This, in turn, altered the costs and benefits facing both the farmers and the king when they entered into a contract. In order to explain the growth in tax farm revenues after 1661, this section presents a simple model that outlines the relationship between the changing transaction costs associated with tax farming, the institutional environment, and the value of a representative tax farm contract.

The expected value of a tax farm contract was probably not identical for all collectors because of differences in knowledge about the tax base or managerial ability. Assume the value of the tax farm to the collector with the highest valuation is,

R(τ,e is the expected revenue of collector i, which increases in the tax rate, τ, and the investment made by the collector in the farm, e. The tax farmer faces a unique constant marginal cost of investment, θ_i. The probability that the collector will lose his capital investment in the farm is given by α. This risk could stem from economic conditions (i.e., a war that reduces trade), nature (i.e., a bad harvest), or capriciousness on the part of the crown (see below).

The investment variable (e) in equation 1 represents any capital that the farmer contributes to the tax farm contract. In 1600 this capital was used both for loans to the crown and investment in the infrastructure and day to day operation of the farm. When first introduced in the fourteenth century, however, the typical tax farm was a small affair and so was the investment required to purchase a tax farm contract. Typical contracts resembled the “1/4 sur les vins en détail” for 1,000 livres, the “1/4 sur les cidres” for 150 livres, or the “1/13 sur les vins en gros” for 60 livres all signed in 1374. Investment in these farms was relatively low because they were simply very small operations. There were no farms in the fifteenth century comparable to the seventeenth-century aides or Gabelles de France, which encompassed many taxes and whose tax base could extend from Burgundy to Normandy.

Given the large number of tax farms and their local nature, the most dominant costs associated with writing a contract during these early years were in allocating the farm to the collector with the highest valuation and extracting as much of that value as possible in the lease price. The competitive English auctions that were used to allocate tax farms until the second half of the seventeenth century were uniquely suited to doing both these things. The collector with the highest valuation for the farm, V₁, wins in an English auction and pays B* (V₂, n), where V₂ represents the second highest valuation and n is the number of bidders. B* (V₂, n) increases in V₂ and n such that

. Simply stated, the auction allocates the farm to the lowest cost manager and, as the number of bidders grows large, the crown appropriates the full value of the tax farm.

The institutions that governed the early tax farm system were designed to ensure the competitiveness of auctions and, as a consequence, also made it costly for farmers to combine their resources by pooling capital. Collectors were discouraged from forming combinations. Leases typically ran for only one year and lessees were not allowed to have partners. An ordonnance of January 1383 relaxed this rule by allowing a limited number of associates if the farm was large enough to justify it. One partner was allowed for a farm worth between 300 livres and 500 livres, two for a farm between 500 livres and 1,000 livres, and three partners if the value was above 1,000 livres.²⁴

These partners were treated as investors with full liability under the law.²⁵ There were also ordonnances stipulating that auctions be publicly announced and allowing government officials to take higher bids up to six months after a lease was signed.²⁶ An ordonnance of 1508 required that farms be leased at the parish level, thereby discouraging unification.²⁷ Officials were forbidden to lease a tax farm to one of their relations, nor were they allowed to lease a farm to, “… any man of the Church, gentleman, or, officer of the king.” This last proscription prevented corruption, on the one hand, because a noble often was the local magistrate. On the other hand, it encouraged competitive auctions because it was within the power of many of the nobility to frighten away potential competitors.²⁸

By the late sixteenth century the institutions that encouraged competitive tax farming were increasingly inconsistent with how tax contracts were evolving. The tax farm system experienced rapid growth from the reign of Henri II. Between 1587 and 1609 the proportion of total receipts represented by the tax farms grew from 34 percent to 72 percent.²⁹

The value of real receipts through the tax farm system tripled between 1600 and 1640.³⁰ This unprecedented growth was, initially, accommodated by increasing the number of tax farms.³¹ However, towards the end of the administration of the Finance Minister Sully, there was a sustained movement towards consolidation of the tax farm system. Between 1605 and 1640 the number of leases signed tended to decline while the value of each lease tended to increase.³² This created incentives for multiple tax farmers to pool their capital. However, because of the rules designed to ensure competition, pooling capital was costly during this period.³³ Thus, individual tax farmers found themselves contributing ever larger amounts of capital to the contracts.

An inevitable by-product of this consolidation was that the amount of capital required to purchase a tax farm lease increased. Why did finance ministers want to consolidate the farms? The first reason is because by 1600 there were economies of scale in tax farming. The second reason is that tax collectors were also increasingly expected to make loans to the crown as part of the initial lease price. Both of these trends altered the expected value of a tax farm contract. Economies of scale meant that individual farmers had to contribute more investment. Larger loans to the crown increased investment while also increasing the likelihood that the king would renege on a contract in order to avoid full repayment. In terms of equation 1, by the beginning of the seventeenth century both α and e were much larger. This resulted in a lower expected value for farms, depressed lease prices, and resulted in lower receipts through the farms.

There are two types of evidence for scale economies in the tax farm system. The first stemmed directly from the decentralized and competitive nature of the early tax farm system. At the end of the sixteenth century the crown had a very tenuous hold on the activities of the tax farmers. In his memoirs Sully recounts how it was common practice for tax farmers to bribe a minister in exchange for a discount on a lease and then resell the lease to other farmers for a profit.³⁴

The result was that the crown did not even have a clear idea of who was collecting its taxes and the farmers often overstepped their authority. Evidence of corruption shows up in royal ordonnances and in complaints made against the farmers by the Estates General of 1614.³⁵ This corruption is consistent with Adam Smith's observation that revenue farmers, “… have no bowels for the contributors, who are not their subjects, and whose universal bankruptcy, if it should happen the day after their farm is expired, would not much affect their interest.”³⁶ Between 1590 and 1653 only about 23 percent of the tax farmers signed more than one contract.³⁷ The vast majority of these signed only two or three. Revenue farmers at the beginning of the seventeenth century did not have much interest in the long-run sustainability of the taxes they managed.

A serious problem was that it was not clear that most finance ministers had “the bowels” for the people during the early seventeenth century either. Under the decentralized tax farm system it was all too easy for a short-sighted finance minister to raise quick cash by creating a new farm and selling it to an equally short-sighted tax farmer. This had the effect of raising the marginal tax rate above its revenue-maximizing value. In effect, under decentralized tax farming, there were too many, independent, and corrupt tax agents exploiting the same common resource.³⁸

For example, in 1595 when Henry IV needed funds for his war against Phillip II of Spain, Sully created a new farm, the “Douane de Vienne.” It was responsible for collecting taxes on, “… all merchandise, be they foreign or otherwise, coming or going, by water or by land …” throughout much of the south of France.³⁹ The Douane had a depressing effect on the value of the existing farm of the Cinq Grosses Fermes, which taxed much of the same trade. This effect was large enough that in 1611, when the farmer La Mothe submitted his bid for the lease on the Cinq Grosses Fermes, it was conditional on the elimination of the Douane de Vienne. However, other farmers were anxious to reinstate the tax; in 1621 it was reinstated, only to be suppressed again at the request of the Cinq Grosses Fermes again in 1624. Finally, in 1626 the Douane was reinstated for the last time before the reforms of Colbert made it a permanent component of the Cinq Grosses Fermes.⁴⁰ The combination of a short-sighted crown leasing tax collection to many, highly independent, short-sighted agents was that the marginal tax rate was too high and trade was diminished. Reducing the number of tax authorities would create incentives more conducive to reducing corruption and lowering the marginal tax rate, as Colbert's reforms would demonstrate.

The second source of scale economies stemmed from the replication of tasks across similar farms. For example, in 1598 the provisionment and sale of salt for the Grandes Gabelles were sold as two separate farms. This caused inefficiencies, as recognized by Sully, who in 1598 unified the two, yielding a price for the combined Grandes Gabelles greater than the sum of its previously separated components.⁴¹

Similarly, between 1614 and 1617 the farm of the Paris Octrois changed hands 15 times. Each of the new farmers brought in his own staff to collect the tax. The situation was so inefficient that the crown issued an ordonnance requiring the farmers of the Octrois to use the same collectors.⁴²

As early as the end of the reign of Henri II, Finance Ministers recognized that there were significant gains to be had by centralizing the tax farm system. The stability of these combinations, however, depended upon the resoluteness of the Finance Minister to protect the property rights of the farmers. Unfortunately, this was increasingly unlikely given the second reason for the centralization of the farms, their increased use as financial intermediaries.

There were two ways in which a tax farmer acted as an intermediary for the crown.⁴³

The most straightforward was for him to personally forward payments on the lease price to the king before they were due. These were called anticipations on future tax receipts. The second, slightly more complicated way, was called an assignation. An assignation was made when a third party (neither the king nor the tax farmer) agreed to lend the crown funds. If the king chose to assign repayment of that debt from tax farm revenues, then the tax farmer could either refuse to buy the debt, buy it at discount, or buy it from the third party at full value. If the debt was purchased, this was called an assignation, the king now owed repayment to the farmer.⁴⁴ Anticipations and assignations were, in effect, loans made by the farmer to the king. They were usually repaid as a charge against the farmer's contracted lease payment. In other words, the farmer was “paid” for his loan through a reduction in one of his future liabilities, the lease payment owed the crown.

In 1623 at least 1,300,000 livres were borrowed through the aides in anticipation of annual tax revenues worth 2,300,000 livres.⁴⁵

The king was anticipating about half of his receipts from the aides before they were collected. In a like manner, the king borrowed at least 1,668,212 livres from the tailles in 1688 in anticipation of future payments of about 2,000,000 livres.⁴⁶ Figure 2 shows the real value of loans as a proportion of annual lease price through the three largest tax farms between 1614 and 1647.⁴⁷

When tax farming was first introduced in France to finance the Hundred Years War, the investments required in the farms were relatively small. However, as the seventeenth century progressed the amount of capital required from the farmers was significant. To provide this capital either multiple farmers could pool a relatively small amount of capital, or a single farmer could provide a large amount. Both of these strategies, however, were in direct conflict with existing institutions that were designed to encourage competition in the tax farm system. Even worse, at the same time that the farmers were trying to bend the rules of competitive tax farming so as to pool their capital, the crown was confronted with greater incentives to “renegotiate” contracts ex post to his advantage. Unfortunately, with the exception of Sully, and later Colbert, seventeenth-century Finance Ministers were not the most resolute bunch. A typical example of how the farms fell apart is illustrated by the case of Antoine Feydeau who, between 1619 and 1623, managed to unify most of the aides and gabelles under his control.⁴⁸

Along with his lease payments he made significant loans to the crown.⁴⁹ In 1623 the crown “renegotiated” Feydeau's contract by refusing to allow him to redeem the full value of his loans through charges. As a result, Feydeau could not meet his lease payments and by 1626 was forced to flee the country. The crown ended up confiscating all of his assets.⁵⁰

In her research on early-seventeenth-century finance Françoise Bayard reveals the many ways the crown violated the property rights of his farmers. For example, the crown could accuse a lease holder of being a “monopolist” so as to get him to “voluntarily” sell his contract to another farmer who was willing to pay more. This was done to the lessor of the aides in 1605 and 1653 and to the Gabelles de Languedoc in 1611.⁵¹

The crown would sometimes accept bids months after a contract was signed. This happened to Jean Palot who took the Convoi de Bordeaux at auction in December 1617. His lease started on 1 July 1618 but on 5 October 1618 a higher offer was made.⁵² The crown could either refuse, accept, or simply wait when these offers were made. Regardless of the choice, however, the effect was to increase the risk surrounding tax farm contracts.

The crown would also often alter the terms of a contract. For example, the number of taxes and the rates at which they were collected could be altered. Sometimes a farm was split up in the middle of a lease. Other times the crown would demand more advances from the farmer than were contained in the original contract. If these alterations were serious enough, the result was often a “rupture” of the tax farm contract. These ruptures were common occurrences during the first half of the seventeenth century. Out of 68 tax farms, 43 experienced at least one rupture. For many, the probability that the contract would not run its full term was significantly higher. Table 3 shows the frequency with which leases were ruptured on the tax farms between 1598 and 1655.⁵³

For example, contracts for the aides were among the most frequently denounced by the king. Between 1598 and 1655, 64.7 percent of them were ruptured. Twenty-two of the 68 tax farms experienced default rates between 30 percent and 50 percent.

What was the effect on the king's revenue of the increased risk of investing in tax farm contracts? If the amount of investment (e) required of the tax farmer was small, then the answer is “probably not much.” In a world of perfect and complete information, the risk created by poor property rights over tax farm contracts would have had no effect on revenue. Bids would decrease to reflect the expected probability of a ruptured contract, but if the king violated contracts half the time, then he would simply collect half as much, twice as often.⁵⁴

By 1600, however, it was necessary to pool capital in order to manage the farms at the efficient scale and finance ministers found this capital too be too tempting a target to leave alone.

The increased fiscal pressure felt by the crown during the seventeenth century significantly altered the costs of writing and enforcing tax farm contracts. The institutions inherited from the fifteenth and sixteenth centuries, before the intense fiscal pressure, were designed to ensure that the tax farms were competitive. By the seventeenth century there was a tension between the incentives of the crown to maximize its short-run payoff by predating on the farms and the increasingly important role played by the farmers as financial intermediaries and investors in the king's tax system. It was not until the second half of the century that this mismatch between institutions and incentives began to change.

INSTITUTIONAL CHANGE

During the first half of the seventeenth century, tax farm contracts were undervalued because of the incompatibility of the existing organization of competitive tax farming with the incentives of the crown to repudiate contracts. The formal institutions that supported competition in the tax farm system made this problem worse because the decentralized tax farmers faced high costs in organizing so as to spread their risk, increase their access to capital, and raise the cost to the king of predating by threatening punishment. In 1661 French fiscal institutions began to adjust to this situation. J. B. Colbert, Louis XIV's finance minister, counted as an integral part of his mercantilist program the consolidation of the tax farm system. In 1663 he negotiated a lease that unified the all of the aides. In 1664 he did the same for the gabelles and traites. Finally, in 1668 he created a truly unified “General Farms” lease which combined all of the major tax farms. Along with consolidation he took the opportunity to “rationalize” the tax rates and rules associated with the farms.

For the customs duties this meant eliminating many of the tariffs associated with the Cinq Grosses Fermes and lowering the remaining taxes across the board. He wrote that, “… the most solid means and most essential for the reestablishment of commerce is the diminution and ordering of taxes which are levied on all merchandise entering and exiting the Kingdom.”⁵⁵

The use of acquits-à-caution, which allowed goods moving across the country to avoid paying multiple taxes, were encouraged. Entrepôts where established where goods could sit for up to six months without being taxed. As mentioned previously, the price of grain decreased significantly after these reforms.

The local taxes associated with the aides, known as the octrois, offer an example of how Colbert created more well-defined and enforced property rights to taxation. The Estates General of 1614 gave towns the authority to manage their individual octrois and keep the resulting revenues. In 1657 Mazarin placed these taxes back under the control of the state. To provide for the fiscal needs of the towns however, he doubled the rates on these taxes with the intention of providing them with the product of this increase. In practice, however, the tax farmers were so far beyond the control of the crown that they largely ignored this provision. As a result, local municipalities resorted to seigniorage. After the consolidation of the aides in 1663 the tax money from the octrois actually went to the towns and this seigniorage stopped.⁵⁶

In the gabelles, Colbert lowered the price of salt, and the unified lease he signed in 1664 ensured that production and sale were contained in the same farm. He made both the aides and gabelles easier to collect by delineating the jurisdiction of the tax farmers more clearly. Court cases concerning farmers were no longer allowed to be tried in seignieurial courts. They had to be brought before royal courts.⁵⁷

Finally, Colbert issued ordonnances that required that the farms not be broken apart. In May 1680 there was an ordonnance for the gabelles.⁵⁸

In June 1680 he issued two for the governance of the aides.⁵⁹ The first covered the jurisdiction of the Cours des Aides in Paris and the second the Cour des Aides of Rouen. These culminated in May 1681 with the issuance of the “Common Title For all the Farms,” which concerned the governance and operation of the unified Company of General Farms and explicitly acknowledged its status as a “legal person.”⁶⁰

Why did not Colbert's push to consolidate the farms end in royal predation, as it had for the previous 60 years? Initially, the unified tax farms succeeded because the farmers trusted Colbert to protect their property rights. Through the judicious use of a Chambre de Justice, Colbert was able to create an informal cartel within the tax farm system based on his personal connections.⁶¹

Because of their close ties with the Finance Minister, the nascent General Farmers were willing to pool their capital without fear of losing their investments like others had before.⁶² Colbert also made it clear to Louis XIV, that he placed a high priority on protecting the property rights of the farmers. In a letter to the king of 11 June 1674 he writes, “I feel obliged to inform you that in order to maintain these farms in this [unified] state so that the farmers can pay the prices of their leases, it is necessary that Your Majesty not only continue to protect the farmers as you have done, but to redouble your efforts …”⁶³ Colbert's steadfastness, combined with a relatively long period of peace between 1661 and 1687 resulted in a stable investment environment for the farmers.

The question remains, however, of why the unified General Farms did not disintegrate after the death of Colbert in 1683? The answer is that, once the initial costs of getting the farmers to pool their capital were overcome, they became “too big to fail.” Under Colbert's protection, the tax farmers became one of the most important financial intermediaries for the crown. After 1687, when the king “assigned” his debts to be paid by the farmers, the company would often convert this debt into short-term paper known as “billets des fermes.”⁶⁴

These billets des fermes were sold to private citizens who could trade them on an active market. The king still owed the company payment on the assignation, but so long as people believed that the company's credit, which was based on the expected tax revenues from the farms, was good, they were willing to buy the billets des fermes. As a result, the company could sustain a much larger amount of lending to the crown than previous, short-lived tax farms could, because it could call on the credit of the broader public. In 1697 there were at least 50,683,407 livres (about one year's lease price) of outstanding billets des fermes.⁶⁵ In real terms, the value of the outstanding billets des fermes in 1697 was equal to the sum of all the lending through the three largest farms between 1600 and 1640.⁶⁶ During the war of the Spanish Succession (1701–1714) the company was so integral for the maintenance of the royal debt that the crown went out of its way to ensure that the farms did not disintegrate, even to the extent that it agreed to pay the farmers a wage in order to protect them from the risk associated with collecting the taxes during wartime.⁶⁷

There are several reasons why the tax farmers were able to pool this capital and use it more effectively at the end of the seventeenth century than at the beginning. First, whereas the General Farms was technically a partnership in which liability was shared equally among the participants, by the end of Colbert's tenure it acted as something more resembling a corporation in which liability was limited.⁶⁸

The last tax farm contract that bore the name of an actual tax farmer was the Lease Legendre in 1668. After that date, the adjudicataire-général, whose name actually appeared on the contract with the king, was usually the valet de chambre of the Finance Minister.⁶⁹ This gave the General Farms a collective personality at law and allowed them to raise capital more easily based on the reputation of the company rather than that of an individual financier. After 1668, when the General Farms made a loan to the crown, it was not treated as debt owed to any individual farmer. If that debt needed to be refinanced, the negotiation occurred between the Company of General Farms and the crown. Thus, unlike the situation in 1623 when Antoine Feydeau lost his personal fortune and was forced to flee the country because of the crown's manipulation of debt, when the crown owed the tax farmers a huge sum in 1697 they were able to negotiate as a group so as to receive payments on it well into the next lease, even though the participants in that lease were not all the same as those who made the original loans to the crown.⁷⁰ The company was also often able to negotiate concessions from the crown when it felt the risk associated with the farms was too high. For example, in 1703, when they refused to sign a lease on the Cinq Grosses Fermes but, instead, required a fixed fee from the king to manage the tax.⁷¹

In order to effectively negotiate with the crown, the company had to prevent one or several of its members from defecting and “going it alone.”⁷²

One way to do this was to restrict membership. After 1683 there are strong indicators this is exactly what happened. Figures 3 and 4 give some idea of the openness of the General Farms by indicating how many farmers participated in multiple contracts before and after 1681.⁷³ Between 1661 and 1680 many different farmers participated in tax farm contracts. Most still only signed one or two contracts. By contrast, after 1681 there is a dramatic increase in the number of farmers who become, in effect, permanent members of the Company of General Farms. The same names appear over and over again on contracts signed during this later period.⁷⁴ At the same time as this shift occurred, the General Farmers also unified their affairs under the same roof. They purchased the Hôtel des Fermes, which remained the center of the tax farming world until the revolution. In the absence of their protector, these shifts can be interpreted as an attempt to lower the costs of acting as a unified group when negotiating with the crown.

Perhaps the most striking evidence concerning the security of the tax farmers' property rights after the reforms of Colbert comes from the two Chambres de Justice that occurred during the second half of the seventeenth century. Chambres de Justice were frequent occurrences during the seventeenth century.⁷⁵

When the crown found itself in financial difficulty, particularly after fighting a war, it would convene a Chambre in order to try the “corrupt” financiers to whom it owed money. Given the disorganized and personal nature of royal finance, especially during the first half of the century, it is difficult to say whether or not a financier brought to trial really had cheated the crown. Nonetheless, in many cases, the Chambre was just a pretext for the crown to default on a loan. At the end of the seventeenth century there were two Chambres de Justice. One in 1661–1665 and the other in 1716. The first was brought by Colbert against the man who preceded him as finance minister, Fouquet. The second was convened after the enormous debt of the crown was revealed in the aftermath of the War of the Spanish Succession. In 1661 tax farmers figured prominently amongst those that were tried and punished. They were assessed significant fines and these fines were actually collected. By contrast, in 1716 members of the Company of General Farms were tried but rarely implicated. The financiers who were severely punished by the Chambre of 1716 were those without connections to major fiscal organizations such as the company.⁷⁶ Furthermore, to the extent that members of the company were fined, or their debt “renegotiated,” these actions were not nearly as enforced as they had been against the farmers in 1665. Indeed, in March 1717 the crown issued a declaration that dismissed the tax farmers, along with other politically well placed financiers, from any further investigations.⁷⁷

Despite these changes, the period between the death of Colbert in 1683 and the fall of John Law in 1720 was one of trials for the General Farms. One the one hand, the farmers were often able to protect themselves from the actions of the state and maintain control over the unified tax farm system. On the other hand, there is no denying that the General Farms were not as powerful in 1683 as they would become by the mid-eighteenth century. War, particularly the War of the Spanish Succession, put tremendous strains on the tax system. After the death of Louis XIV in 1715 it was revealed to the public that the government had a total debt of 2 billion livres, which it was supporting with about 600,000,000 livres worth of short-term paper, much of it billets des fermes owed to the company. The entire tax system, by contrast was only capable of producing some 165,000,000 livres a year.⁷⁸

The Chambre de Justice of 1716 was instituted in as a strategy to manage this debt. Despite the fact that the farmers escaped the Chambre without great loss, a much more serious threat from across the Channel arrived on the scene soon after.

In 1716 the Scotsman John Law received permission from the Regent to implement financial reforms that bore many similarities to the innovations England had adopted 18 years earlier. He created a central bank with assets based on stock floated by a joint stock company with a monopoly on overseas trade. In May 1716 the bank started taking depreciated government paper in exchange for shares. More importantly, from the point of view of the General Farms, Law declared the bank's notes legal tender in payment of taxes.⁷⁹

This led a group of financiers, among them many of the General Farmers, to counter Law's System with their own “Anti-System.” They signed a new lease on the General Farms and created their own joint-stock company that raised capital equal to that of Law's bank. Then, they began redeeming the millions of livres worth of bank notes that had been forced upon them, in the process placing overwhelming demands on the gold reserves of Law's bank.⁸⁰ This forced Law in 1719 to convince the government to allow him to bring the tax farms, along with the rest of the tax system, under the control of his Indies Company. With this move, Law possessed a monopoly over the colonial trade, was the sole collector of taxes, minter of money, and was on his way to converting all of the royal debt into Company of Indies stock. In 1720, the bank and the Company of the Indies were unified and Law became the controller general of the system.

By the end of 1720 the public had lost confidence in the inflated share prices for the Company of the Indies, the system was in shambles, and Law was an exile. Within a year of Law's demise, in a rush back to the status quo, the crown began the process of reinstating the Company of General Farms into its previous role. Why exactly Law's reforms failed is a subject of debate. However, one recent study acknowledges that an inherent contradiction in the system was that, “… Law was both the king's minister and the Company's CEO, and ultimately, the Company's powers and monopolies derived from the king's will.”⁸¹

It is clear that Law manipulated the price of shares in the company as early as 1718, ultimately leading to the conditions responsible for the collapse in 1720.⁸² We will never know exactly why he did this, but with the crown pressuring Law to convert the government debt on as favorable terms as possible, a conflict of interest certainly existed between Law's obligations to the stockholders of the Company of the Indies and his obligations to the king.

Law's system suffered from the same fundamental flaw as did the competitive tax farming system of the early seventeenth century. Its organization was not consistent with the institutions of absolute monarchy. Law created fiscal organizations in France, like a joint-stock company and central bank, that worked fantastically well in England. The difference was that England had a parliament whereas France had an absolute monarch. Just as the crown faced no constraints from predating on the early, unorganized, tax farmers, Law faced no real constraints in manipulating his system, at the cost of unorganized stock holders.⁸³

Given the lack of formal political constraint against the absolute monarch, the Company of General Farms represented an organizational form that was superior to Law's because it had the ability to protect its property rights while at the same time keeping the interests of its investors separate from those of the crown.