Explaining financial market puzzles with learning

Asset prices often do not seem to follow the predictions of standard economic models that assume investors possess fully rational expectations. For example, such models suggest (1) that basic arbitrage relationships in foreign exchange markets are consistently violated (the “uncovered interest rate parity puzzle”); (2) that stock prices are excessively volatile given the volatility of fundamentals (the “excess volatility puzzle”), and (3) that the historical returns earned on equities are far too high in comparison with the returns on risk-free assets (the “equity premium puzzle”). This article examines the implications that imperfect knowledge and learning have for the determination of exchange rates and equity prices. It illustrates how the above-mentioned empirical phenomena can be reconciled with efficient and well-functioning asset markets in which investors possess small amounts of imperfect knowledge. JEL Classification: G12