We consider to what extent the empirical failings of the Q model of investment can be attributed to the use of share prices to measure average q. We show that the usual empirical formulation may fail to identify the Q model when stock market valuations deviate from the present value of expected net distributions in ways that are consistent with weak and semi-strong forms of the Efficient Markets Hypothesis. We show that the structural parameters of the Q model can stil be identified in this case using a direct estimate of the firm's fundamental value, and implement this using data on securities analysts' earnings forecasts for a large sample of publicly traded US firms. Our empirical results suggest that stock market valuations deviate significantly from fundamental values. Controlling for this, we find no evidence that the Q model of investment is seriously misspecified.