A test of the relationship between financial asset values and the fundamentals is presented. The test is an indirect test based on Tobin's (1969) q model of investment. The advantage of this test is that it avoids conditioning on a specific model of equilibrium returns by substituting an observable proxy, capital, for the unobservable fundamental value of the firm. It is shown that under mild restrictions, the value of capital and the financial value of the firm should move together even if marginal or average q does not accurately describe the actual investment decision rule. Average q is a mean reverting process. The results of tests for violations of the mean reversion restriction are presented. The tests cannot reject the unit root null hypothesis for a long annual sample from 1926 to 1988 or a quarterly post-World War II sample. The power of the tests is examined with Monte Carlo simulations. The simulations show that the tests have good power until the serial correlation in the series gets quite high. The tests cannot distinguish between no mean reversion and very slow mean reversion. The empirical evidence presented in this paper provides fairly strong evidence against the hypothesis that the financial and real values are closely linked in the short run or even in the medium run.