An Asset Market Integration Test Based on Observable Macroeconomic Stochastic Discount Factors

There are a number of tests and measures of the degree of integration in the literature. An example is the idea that integrated markets should provide rates of return that are highly correlated with one another and that a measure of correlation provides an appropriate test. This particular idea is clearly false; for substantial periods of time we don’t ever see stocks traded on the same market moving together. Specific models of what prices risk in individual markets could provide the basis of a test of integration. However, as has been widely shown, any differences between these pricing models will be subject to arbitrage by informed traders and so cannot form the basis for a test. In this paper we exploit the absence of arbitrage possibilities and the operation of the ’Law of One Price ’ in stochastic discount factor (SDF) theory to construct a test of integration based on a common approach to pricing assets in all markets, not only for stocks. The SDF approach that we adopt says that one SDF should price all assets as the model is not market or asset-specific.Unlike much of the literature, we adopt a direct parametric approach which takes estimates of an identical SDF from two asset markets and asks whether the price of risk associated with this SDF is th