This paper analyzes the relationship between exchange rates and economic fundamentals in a rational expectations present value model. It shows that when fundamentals are I(1) processes and the discount factor is near one, asset prices exhibit near random walk behavior. This result helps explain why fundamental variables such as money supplies, outputs, inflation, and interest rates provide little predictive power for exchange rate changes. However, the data do show a related link suggested by standard models - that exchange rates help predict these fundamentals. The implication is that exchange rates and fundamentals are linked in a way consistent with asset pricing models of the exchange rate. The paper presents a theorem showing that in a present value model, asset prices follow a process arbitrarily close to a random walk if (1) at least one forcing variable has a unit autoregressive root, and (2) the discount factor is near unity. This result is not an application of the simple efficient markets model of Samuelson, but rather a more nuanced analysis of how discount factors and fundamentals influence asset prices. The paper also demonstrates that exchange rates can help predict fundamentals, especially nominal variables. Using quarterly bilateral dollar exchange rates, 1974-2001, for the dollar versus the six other G7 countries, the authors find some evidence of such causality. The statistical significance of the predictability is not uniform, but the results suggest a link between exchange rates and fundamentals that is modest compared to other economic variables. The paper also examines the implications of these findings for exchange rate models. It shows that in the class of models considered, all empirical results are consistent with at least one other explanation - that exchange rate movements are dominated by unobserved shocks that follow a random walk. The plausibility of this explanation is underscored by the fact that the authors generally fail to find cointegration between the exchange rate and observable fundamentals. The paper concludes that while exchange rates may follow a random walk, they can still help predict fundamentals. This is because the random walk behavior is due to the discount factor being near unity, which makes the exchange rate a poor predictor of fundamentals. However, the paper also finds evidence that exchange rates Granger-cause fundamentals, suggesting that exchange rates can be useful in forecasting these variables.This paper analyzes the relationship between exchange rates and economic fundamentals in a rational expectations present value model. It shows that when fundamentals are I(1) processes and the discount factor is near one, asset prices exhibit near random walk behavior. This result helps explain why fundamental variables such as money supplies, outputs, inflation, and interest rates provide little predictive power for exchange rate changes. However, the data do show a related link suggested by standard models - that exchange rates help predict these fundamentals. The implication is that exchange rates and fundamentals are linked in a way consistent with asset pricing models of the exchange rate. The paper presents a theorem showing that in a present value model, asset prices follow a process arbitrarily close to a random walk if (1) at least one forcing variable has a unit autoregressive root, and (2) the discount factor is near unity. This result is not an application of the simple efficient markets model of Samuelson, but rather a more nuanced analysis of how discount factors and fundamentals influence asset prices. The paper also demonstrates that exchange rates can help predict fundamentals, especially nominal variables. Using quarterly bilateral dollar exchange rates, 1974-2001, for the dollar versus the six other G7 countries, the authors find some evidence of such causality. The statistical significance of the predictability is not uniform, but the results suggest a link between exchange rates and fundamentals that is modest compared to other economic variables. The paper also examines the implications of these findings for exchange rate models. It shows that in the class of models considered, all empirical results are consistent with at least one other explanation - that exchange rate movements are dominated by unobserved shocks that follow a random walk. The plausibility of this explanation is underscored by the fact that the authors generally fail to find cointegration between the exchange rate and observable fundamentals. The paper concludes that while exchange rates may follow a random walk, they can still help predict fundamentals. This is because the random walk behavior is due to the discount factor being near unity, which makes the exchange rate a poor predictor of fundamentals. However, the paper also finds evidence that exchange rates Granger-cause fundamentals, suggesting that exchange rates can be useful in forecasting these variables.