Conventions and the stock market game
Forecasting stock price movements is a notoriously difficult job. Were it not so, it would be easy to get richer. In this case, however, nobody would get poorer. But if nobody gets poorer, nobody will get richer. There are two ways to get out of this vicious circle. The first, and the more well-trodden, is the Efficient Market Theory (EMT), or: Everybody Understands Everything. The second is the Casino Market Theory (CMT), or: Nobody Understands Anything. This work is an attempt to bridge the gap between these two theories. In the first chapter the EMT is analysed in its fundamental constituents, while Chapter 2 contains a discussion of several empirical tests of the theory. Chapter 3 extends the EMT to incorporate variable risk premia and rational speculative bubbles and Chapter 4 presents the available empirical evidence on the extended model. The line of research based on the EMT paradigm is abandoned in Chapter 5, where the central principle of the EMT - the assumption of homogeneous investors with common priors - is investigated and challenged. The basis is there laid for an alternative view of the stock market game, which emphasises the conventional nature of investors' beliefs about future returns and is consistent with the view that stock market prices do not only reflect the fundamental value of underlying companies. In Chapter 6, the hypothesis that non fundamental information (in particular, past information) may have an influence on current stock prices is evaluated against monthly data relative to the US, UK, Japanese and Italian stock markets. Contrary to popular wisdom, we find that past information has a significant effect on current stock returns. Our evidence indicates that, as Keynes suggested in the General Theory, conventional beliefs play a crucial role in the stock market game.