In the economist, Nobel laureate Bob Lucas writes on the economic profession and discusses the ideas of efficient markets. Coffeehouse economics regulary puts the idea of efficient markets on the agenda, given its pivotal role in thinking about market dynamics, in particular on financial markets.
To me, the most intuitive way of thinking about the hypothesis of markets being efficient is that in its strict form it means that there is no possibility of achieving a risk-free future profit with a zero net investment. If a trading strategy of this kind were possible – in other words, if the possibility of arbitrage existed – the price adjustments resulting from the trading activities would eliminate the arbitrage opportunity. Consequently, the concept is a equilibrium definition as put forward by amv like many other equilibrium definitions in economics.
In this respect, Bob Lucas points out that
The term “efficient” as used here means that individuals use information in their own private interest. It has nothing to do with socially desirable pricing; people often confuse the two.
Indeed, empirical research indicates to highly efficient markets, if one accepts above definition. It is for that reason why, from a macroeocnomic perspective, bubbles may emerge although financial market are still efficient. Financial assets are correctly priced since market participants "use information in their own private interest". For a further discussion readers of the coffeehouse may be referred to Lucas' roundtable.
Remark: I do not aggree with all the issues, Lucas discusses in his op-ed, in particular the forecasting power of all these fancy DSGE models. But I think, he makes a good point in defence of the no-arbitrage approach, which is heavely rejected by many, many economists.
