The discovery of America was followed by a flow of precious metals to Spain and Portugal, and from there throughout the world. Historiography has reconstructed the quantities of gold and silver transferred from the New World to the Old World in the Early Modern period, but what was the reason for the bullion outflows? This dissertation answers this question. In particular, it examines the logic of silver outflows from Cadiz to London in the first half of the 18th century.Castile enacted bullionist laws during more than four centuries, from the Late Middle Ages to mid-19th century. The laws fixed prices and placed bans on export. But these measures did not prevent the export of silver and instead caused a great deal of smuggling. This dissertation aims at understanding the logic of silver outflows focusing on the smugglers' point of view: arbitrage. In this regard, the archive of the merchant house Roux (Marseille), probably the best preserved 18th century commercial archive in Europe, has made possible the reconstruction of the specie-point mechanism for silver - the Old Mexican pieces of eight - between Cadiz and London as exactly practiced by contemporary merchants. The discovery of half-monthly data on silver black market in Cadiz for the period 1729-1741 has been a milestone in order to understand the logic of silver outflows.Empirical result from these data for arbitrage equation presents a puzzle for our understanding of the specie-point mechanism: from 1729 to 1737 there was a systematic bias between the implicit spot exchange rate and the arbitrated parity, which made arbitrage systematically profitable. On the contrary, from 1737 to 1741 the bias was corrected because the Spanish government reacted to illegal bullion outflows with a devaluation, which equalized the exchange rates and the arbitrated parity.This research explores both theoretically and empirically the reasons for the apparent mispricing for the first period and the effect of the devaluation on silver prices for the second period. The outcome is that bullionist regulations configured an oligopsony structure in Cadiz that had the power to drive down silver prices below the international price (i.e., London price). Oligopsony agents were the most important foreign merchants in Cadiz, organized in family and partnership networks which were rice-makers; their structure was maintained because the long-run international networks created entry barriers in the business of illegal export of bullion. Secrecy was reserved because both sides of the market cheated the Spanish government: importers from the Spanish American colonies saved the high import tax and exporters to the ain European bullion markets ignored the ban against exports.Nevertheless, oligopsony power had a floor, which was the Official Parity (i.e., the number of units of account per coin). Below the Official Parity, the pieces of eight were used as money and went out from the commodity market. The devaluation of 1737 should be understood as an increment of the Official Parity for eliminating oligopsony power.Some main lessons emerge from this dissertation. First, understanding the reasons of the specie flows in the Early Modern period demands comprehension of the specie-point mechanism. Second, the construction of the silver-points requires the location, collection and manipulation of the right data: market prices, exchange rates and costs of arbitrage. And third, the interpretation of the arbitrage results needs to focus on the special microeconomic features of the bullion market structure. This is an original approach which will provide a lot of insight into the workings of commodity money.The first chapter describes the Castilian stagnated legislation and immobile institutions established with the aim of avoiding bullion outflows: fixed prices and bans on export. The second chapter analyses the specie-point mechanism in the institutional setting of bullion controls: the case of silver Pieces of Eight between Cadiz and London during the period 1729-1741. Arbitrage equation shows a systematic bias between the spot exchange rate and the arbitrated parity corrected by the 1737 devaluation. The third chapter analyses the specie-point mechanism in the institutional setting of free bullion movement: the case of gold and silver bars between London and Amsterdam during the period 1734-1758. London-Amsterdam bullion market was integrated, and arbitrage equation shows only few and non persistent breaks. The fourth chapter tells the story of the agents involved in the illegal exchanges of silver in Cadiz, and demonstrates that the smugglers were the French merchants who obtained the highest income of all merchants in Cadiz. The fifth chapter examines the contemporary Castilian reports against smuggling in order to describe how the illegal exchange took place. Smuggling was reserved to foreign merchants because they had achieved privileges which prevented them to be prosecuted. The sixth chapter demonstrates that the smugglers were organized in long-run networks which conferred them the market power to drive down bullion prices below the international price, and the international connections to illegally extract and distribute the bullion from Cadiz. The seventh chapter develops a static model of partial equilibrium for commodity-money in order to understand the workings of the oligopsonistic silver-commodity market and the effect of devaluation on the bullionist goal of treasuring silver. We will end offering some conclusions. Appendices explain the construction of the specie-point mechanism.