In my finance class we are learning about efficient market theory. There are three versions of this theory:
Weak - All historical data is priced into the market
Semi-Strong - All historical and publicly available information is priced into the market
Strong - All information both public and private is priced into the market
The Weak theory implies that it is not possibly to beat the market by looking at past historical data (apparently there is some academic dispute regarding momentum but I digress)
The Semi-Strong implies that it is not possible to beat the market utilizing all past and present public information
The Strong means that you can't beat the market. Period.
Then we talked about Perfect markets. A perfect market is one that is efficient AND has no transaction costs. An efficient market is not necessarily perfect.
Anyway this is the less interesting part of my diatribe, the interesting part is the question the professor asked: If we are telling students that they should pursue positive NPV projects for their future companies but you can't beat the market then isn't this a contradiction? I admit to not knowing the answer (I hate when that happens) but the answer is that you are talking about two different markets. The financial market is what is generally referred to as the market. The real market is when you actually take resources, lumber and land, and turn them into more productive assets, a house. THAT market is not efficient (or not nearly as efficient as the financial market) and there are arbitrage opportunities.
To show the inefficiency of the real market he gave an example. Let's say the price of oil goes up, do the plastic spoons sold in the grocery rise immediately in price. No they don't. Eventually they will, but the lag between the rise in the factor of production, which should raise the price of the product immediately, is inefficiency in the real market.