Integration and arbitrage in the Spanish financial markets: An empirical approach*

Several authors have introduced different ways to measure integration between financial markets. Most of them are derived from the basic assumptions about asset prices, like the Law of One Price or the absence of arbitrage opportunities. Two perfectly integrated markets must give identical prices to identical final payoffs, and a vector of positive discount factors, common to both markets, must exist. If these properties do not hold, the degree to which they are violated can be defined and considered as a measure of integration. The present paper empirically tests integration measures in the Spanish financial markets. Furthermore, the integration measures are operationalized to analyze the presence of cross‐market arbitrage without previously specifying the exact nature of the arbitrage strategy to be used. When the absence of arbitrage holds, several interesting variables are obtained, for instance, state prices or risk‐neutral probabilities. When this absence fails, explicit cross‐market arbitrage portfolios are provided. The results of the test yield some evidence about market efficiency and integration outside the United States, and they are surprising for several reasons. First of all, arbitrage opportunities do sometimes appear, and the bid–ask spread and transaction costs seem to be unable to prevent arbitrage profits. Furthermore, the criticisms that are usually raised when empirical papers show the existence of arbitrage opportunities do not apply here because we work with perfectly synchronized high frequency data. On the other hand, different integration measures show a similar evolution along the tested period, although these measures give different information about the markets’ efficiency and integration, and they do not necessarily have to be related. © 2000 John Wiley & Sons, Inc. Jrl Fut Mark 20: 321–344, 2000